Search

ABSTRACT: Increased deposition of hyaluronan (HA) is part of the early response to fibrogenic stimulus in the lung exposed to bleomycin injury and has been associated with increased lung water in adult animals. Early respiratory distress syndrome (RDS) in premature infants is characterized by increased lung water, and late sequelae include fibrosis or bronchopulmonary dysplasia. We hypothesized that increased MA in the alveolar interstitium would be associated with increasingly severe RDS in prematurely delivered monkeys and that modes of therapy that affect severity of disease such as treatment with high-frequency oscillatory ventilation or exogenous surfactant would decrease this response. Thirty-four Macaca nemestrina monkeys were delivered at 134 ± 1 d (term = 168 d) and randomized to high-frequency oscillatory ventilation or conventional mechanical ventilation from birth. Sixteen of these animals received surfactant. At 6 h of age, the right lower lung was frozen in situ during inflation to 30 cm H2O (approximately 2940 Pa) and then dehydrated and processed for microscopy. The presence and severity of RDS were evaluated by clinical and morphologic criteria. HA concentrations in lung extracts increased with progressively severe RDS (p = 0.0003). Treatment with high-frequency oscillatory ventilation decreased the lung injury score (1.69 ± 0.7 compared with 2.5 ± 0.9, p = 0.05), but changes in lung HA concentration did not reach significance (37.9 ± 22.7 compared with 44.8 ± 22.6). Surfactant treatment decreased lung HA concentration (29.6 ± 19.0 µg/wet lung) compared with non-surfactant-treated animals (54.7 ± 20.2 µg/g wet lung, p = 0.0009). Two fetal animals (144 and 163 d gestation) and seven additional premature animals ventilated for up to 96 h were compared with the animals killed at 6 h. HA concentrations increased with length of mechanical ventilation and severity of illness in these animals. HA was localized in freeze-dried lung sections using a biotinylated probe. Lung sections were blindly scored for the distribution of HA staining, and these scores were positively correlated with HA concentration measurements (r = 0.75, p < 0.0001). The quantity of HA in alveolar microvasculature correlated with severity of RDS (r = 0.68, p = 0.0004). We conclude that 1) HA concentration in RDS lungs of prematurely delivered infant monkeys is increased relative to normal lungs at 6 h, 2) increased HA is localized predominantly to the perivascular space of lung vasculature, and 3) this response is decreased by surfactant treatment.

• We reviewed the records of 4778 infants who were admitted to the affilliated intensive care nurseries at the University of Washington In Seattle from Jan 1, 1980, through Dec 31, 1983. We evaluated the outcome for patients with bronchopulmonary dysplasia who required mechanical ventilation and supplemental oxygen for at least six months. Eight (53%) of 15 study patients were alive at 3 years of age. Preselected indexes measured during the first six months of life that were not associated with death before 3 years of age included growth measurements, fraction of inspired oxygen at 6 months of age, mean airway pressure, arterial partial pressure of carbon dioxide, and right ventricular hypertrophy. In contrast, recurrent cyanotic episodes requiring intermittent muscle paralysis or long-term sedation therapy to maintain gas exchange occurred in six of seven nonsurvivors and only one of eight survivors. The survivors have reactive airway disease and recurrent lower respiratory tract infections.(AJDC 1988;142:721-725)

Neonatal group B streptococcal (GBS) sepsis produces pulmonary arterial hypertension and hypoxemia that are preventable by pretreatment with the selective thromboxane A2synthase inhibitor, dazmegrel. In the present experiment we administered dazmegrel (8 mg/kg) 2 h after the initiation of a 2 h infusion of 5 × 108GBS/kg/h in ten 2− to 3-wk-old piglets. The multiple inert gas elimination technique was used to measure intrapulmonary shunt and alveolar ventilation to pulmonary perfusion mismatching. Thromboxane B2, the stable metabolite of thromboxane A2, and 6-keto-prostaglandin F1α, the stable metabolite of prostacyclin, were assayed in arterial blood. Pulmonary arterial pressure increased immediately after initiation of the GBS infusion, rising from 12 ± 2 to 34 ± 4 torr (p< 0.02); pulmonary vascular resistance increased by 400 (p< 0.01). Arterial hypoxemia developed (p< 0.02) in association with an increase in the low ventilation-perfusion ratio index but without a significant increase in intrapulmonary shunt. Thromboxane B2levels increased 10-fold. Infusion of the carrier substance for dazmegrel after 2 h of GBS infusion produced no change in any variables. In contrast, infusion of the drug resulted in the return to pre-GBS infusion baseline values for both pulmonary arterial pressure and pulmonary vascular resistance. However, no improvement in arterial pO2or in the low ventilation-perfusion ratio index occurred. Both pulmonary vascular resistance and pulmonary arterial pressure remained normal for 0.5 h after dazmegrel administration despite continued GBS infusion. Thromboxane B2levels were decreased 30 min after dazmegrel (p< 0.02), but remained greater than pre-GBS levels. Dazmegrel reversed pulmonary hypertension and elevated pulmonary vascular resistance, but did so without concomitant improvement in pulmonary gas exchange, when administered after 2 h of GBS infusion.

Group B streptococcus (GBS), a common neonatal gram-positive pathogen, causes similar pathophysiology in human newborns and neonatal animal models of sepsis. Animal models of GBS sepsis demonstrate a two-phase response: 1) an acute phase (<1 h) of increased pulmonary artery pressure (Ppa) and reduced arterial oxygen pressure (Pao2) that is associated with increased serum thromboxane B2(TxB2) and 2) a late phase (2–4 h) of persistently increased Ppaand reduced Pao2, reduced systemic arterial pressure, and progressive fall in cardiac output that is associated with increased serum TxB2, 6-keto-prostaglandin Fl$$ (6-keto-PGFi$$), and tumor necrosis factor-α (TNF$$). We hypothesized that pretreatment of piglets with both pentoxifylline (PTF), an inhibitor of TNF$$ production and activity, and indomethacin (INDO) would 1) inhibit GBS-induced TxB2, 6-keto-PGF1, and TNF“ and 2) prevent both the acute- and late-phase physiologic responses of GBS sepsis. Combined PTF and INDO pretreatment of anesthetized, mechanically ventilated piglets infused with GBS (1.25 x 109colony forming units/ kg/h) for 4 h prevented GBS-induced increases in Ppa at 1 h (GBS+PTF+INDO: 1.8 $$ 0.07 kPa versus GBS alone: 4.7 ± 0.1 kPa) and markedly attenuated increases in Ppa at 4 h (GBS+PTF+INDO: 2.1 ± 0.1 kPa versus GBS alone: 4.4 ± 0.1 kPa). PTF+INDO treatment prevented GBS-induced reductions in both mixed venous oxygen pressure and Pao2at 1, 2, and 4 h (GBS+PTF+INDO: 11.5 ± 0.4 kPa versus GBS alone: 7.1 ± 0.4 kPa), and attenuated GBS-induced declines in cardiac output. PTF+INDO treatment significantly attenuated GBS-induced serum TNF$$ polypeptide levels (ELISA, pg/mL) at 4 h (GBS+PTF+INDO: 143 ± 45 versus GBS alone: 502 ± 147) and blocked GBS-induced increases in serum TxB2 and 6-keto-PGF1 (all levels < 10 pg/0.1 mL by RIA). INDO pretreatment alone prevented GBS-induced increases in serum TxB2and 6-keto-PGF1$$ levels but did not significantly inhibit GBS-induced TNF$$ production. INDO pretreatment alone did not attenuate GBS-induced increases in Ppa at 4 h, nor prevent late-phase reductions in both Pao2 and mixed venous oxygen pressure. PTF+INDO treatment of GBS sepsis in piglets is superior to treatment with INDO or PTF alone. Inhibition of both blood eicosanoid and TNF$$ production may provide adjunctive therapy for human newborns with sepsis and pulmonary hypertension.

Acute hypoxic pulmonary vasoconstriction is attenuated by respiratory alkalosis. It is unknown if alkalosis similarly reduces pulmonary vasoconstriction produced by thromboxane A2. Respiratory alkalosis does not always attenuate persistent pulmonary hypertension in newborns, some of whom have elevated serum thromboxane B2 levels. We hypothesized that alkalosis attenuates thromboxane-induced pulmonary vasoconstriction less than it does hypoxic pulmonary vasoconstriction in infants. Hemodynamic responses to respiratory alkalosis during pulmonary vasoconstriction produced in random order by breathing 12 inspired oxygen and by infusing 0.1 μg/kg/ min of the thromboxane-mimetic U46,619 were compared in eight 2-wk-old piglets. Hypoxia increased mean pulmonary artery pressure from 12 ± 3 to 29 ± 2 mm Hg and pulmonary vascular resistance (PVR) from 11 ± 4 to 25 ± 8 mmHg/L/min; U46,619 increased pulmonary artery pressure from 16 ± 5 to 37 ± 6 mm Hg and PVR from 14 ± 5 to 51 ± 17 mm Hg/liter/min. U46,619 also decreased cardiac output accounting in part for the greater increase in PVR compared to hypoxia-induced vasoconstriction. Respiratory alkalosis decreased PVR to 14 ± 6 mm Hg/ liter/min during exposure to hypoxia and to 28 ± 9 mm Hg/liter/min during infusion of U46,619. In six additional piglets with U46,619-induced pulmonary vasoconstriction, the effects of lung stretch and hypocapnic alkalosis were separated by doubling tidal volume and then adding inspired CO2to return PaCO2to prehyperventilation levels. Respiratory alkalosis decreased PVR from 52 ± 36 to 35 ± 21 mm Hg/liter/min. Despite the increased tidal volume, PVR increased to 53 ± 35 Hg/liter/min when PaCO2returned to 44 ± 5 mm Hg. Respiratory alkalosis rather than lung stretch reduces thromboxane-induced and hypoxia-induced pulmonary vasoconstriction by equal proportions.

Thromboxane-associated pulmonary hypertension occurs in animals during intravenous infusion of group B streptococcus (GBS), a gram-positive neonatal pathogen. We postulated that other gram-positive neonatal pathogens, such as Streptococcus fecalis (ENT) and Staphylococcus epidermidis (S. epi) would also induce increased thromboxane synthesis and pulmonary hypertension when infused into piglets. We observed similar hemodynamic and gas exchange abnormalities during stepwise increases in the dose of GBS, Ent, and S. epi (n = 3, 4, and 4 piglets receiving each bacteria, respectively). Pulmonary vascular resistance increased significantly in the absence of acidosis or reduced arterial or mixed venous pO2at a dose of 2.5 ± 108cfu/kg/h for Ent and S. epi. In 14 additional piglets, pulmonary vascular resistance increased markedly after 60 min of intravenous infusion of 4 ± 1 ± 108cfu/kg/h for each organism (p < 0.05, GBS: 11.7 ± 1.8 to 75.6 ± 18.4 mm Hg/liter/min, Ent: 12.7 ± 1.7 to 64.9 ± 10.6 mm Hg/liter/min, S. epi: 10.5 ± 0.8 to 56.9 ± 6.0 mm Hg/liter/min), and blood thromboxane B2 levels increased (p < 0.05, GBS: 30 ± 10 to 1830 ± 330 pg/nl, Ent: 20 ± 7 to 1110 ± 300 pg/ml, S. epi: 31 ± 9 to 1260 ± 350 pg/ml). This dose of each bacteria caused a similar degree of mild arterial hypoxemia (57–66 mm Hg). The thromboxane synthetase inhibitor, dazmegrel, completely reversed pulmonary hypertension, reduced TxB2 levels to near baseline values, and partially reversed arterial hypoxemia despite ongoing bacterial infusion. We conclude that thromboxane-associated pulmonary hypertension occurs in piglets during infusion of different gram-positive neonatal pathogens.

ABSTRACT. Group B streptococcal (GBS) sepsis produces arterial hypoxemia in newborns. In piglets we previously found that hypoxemia develops because of increased ventilation perfusion heterogeneity, and reduced mixed venous pO2occurring in association with decreased pulmonary blood flow. We hypothesize that increased thromboxane A2(TxA2) synthesis mediates the immediate alterations in gas exchange found in GBS sepsis. We studied 18 anesthetized, ventilated piglets before, during, and after a 30-min infusion of 2 × 109colony forming units/kg of GBS. Nine piglets were pretreated with 8 mg/ kg of dazmegrel (DAZ), a TxA2synthetase inhibitor, and nine animals received GBS without DAZ pretreatment. Pulmonary and systemic arterial pressures, pulmonary vascular resistance, pulmonary blood flow, respiratory gas tensions, intrapulmonary shunt, and SD of pulmonary blood flow, an index of ventilation perfusion mismatching, were measured. Systemic and pulmonary arterial levels of thromboxane B2and 6-keto-PGF1αwere also measured. The sham-treated animals showed the expected rise in pulmonary arterial pressure from 12 ± 3 to 29 ± 7 torr, (p<0.02). By comparison, the animals pretreated with DAZ did not demonstrate pulmonary arterial hypertension and had a delay in the fall in pulmonary blood flow until 2 h postinfusion. Arterial PO2did not decline significantly after the GBS infusion in the DAZ-pretreated animals; the untreated animals showed a significant fall in pO2from baseline. There was no significant change in intrapulmonary shunt or SD of pulmonary blood flow compared to baseline in the DAZ-pretreated animals. The elevation in thromboxane B2occurring with GBS sepsis did not occur in the DAZ-pretreated animals. We conclude that TxA2in part mediates the immediate gas exchange and pulmonary hemodynamic abnormalities during GBS sepsis. Inhibition of TxA2synthetase results in preservation of normal pulmonary gas exchange and a delay in the fall in Qp following GBS infusion.

ABSTRACT: Lobar atelectasis is common among infants and children with obstructive lung diseases. However, the effect of lobar collapse in the presence of diffuse lung disease on gas exchange in the pediatric age group has not been described. We developed an infant model of lobar atelectasis using piglets and tested the hypothesis that diffuse alveolar hypoxia increases shunt fraction (Q/QT) associated with lobar atelectasis by redirecting pulmonary blood from the well-ventilated portion of the lung into the collapsed region. Shunt fraction was determined using the multiple inert gas elimination technique. The proportion of pulmonary blood flow perfusing the left lower lobe was measured with microspheres. Q/Qtincreased significantly but by a variable amount to an average value (±SD) of 5.9 ± 4.2% following lobar collapse. The percentage of cardiac output perfusing the left lower lobe fell by an average of 70 ± 17% in response to lobar collapse. When animals were ventilated with 12% oxygen, shunt fraction increased to 18.7 ± 7.1% and blood flow to the collapsed left lower lobe increased from 9 ± 3 to 22 ± 3% of cardiac output. Lobar atelectasis in conditions where diffuse alveolar hypoxia is present may be associated with a significantly greater intrapulmonary shunt than lobar atelectasis in children with otherwise normal lungs.

Summary: The possible acute deleterious effects of maternal glucocorticoid administration on the fetal nervous system and the pathologic significance of sudanophilic lipids in glial cells were studied in the premature pigtail monkey (macaca mulatta). At 72, 48, and 24 h before delivery at 135 ± 1 d gestation, dams were treated with either 4 mg dexamethasone or saline. After delivery, respiratory function of each fetus was determined and supported. The animals were sacrificed at 3 h of age.Brain weights were similar in the two groups. Dark, shrunken, pyknotic neurons were present in the hippocampus of three treated and two control animals. Lipid-containing glial cells were present in all animals. Neither appeared to be related to steroid treatment or to the degree of respiratory distress.We conclude that short-term glucocorticoid therapy in doses analogous to those used in humans for the prevention of hyaline membrane disease does not result in acute neuronal damage. Sudanophilic lipid accumulation in glial cells is not always abnormal and must be distinguished from glial fatty metamorphosis.

Group B streptococcal sepsis in newborns produces pulmonary arterial hypertension and hypoxemia. The purpose of this study was to investigate the mechanisms by which hypoxemia occurs. Ten anesthetized, ventilated piglets were infused with 2 x 109colony forming unitskg of Group B streptococci over a 30-min period. Pulmonary arterial pressure rose from 14 ± 2.8 to 38 ± 6.7 torr after 20 min of the bacterial infusion (p< 0.01). During the same period, cardiac output fell from 295 to 184 mlkgmin (p< 0.02). Arterial Po2declined from 97 ± 7 to 56 ± 11 torr (p< 0.02) and mixed venous Po2fell from 39.6 ± 5 to 28 ± 8 torr (p< 0.05). The multiple inert gas elimination technique was used to detect increases in shunt and alterations in ventilation-perfusion matching. Intrapulmonary shunt did not increase during or after the infusion with group B streptococci. However, there was a significant increase (p< 0.05) in the SD of pulmonary blood flow, an index of Amismatching, 20 min after initiation of the infusion of bacteria. All the above changes reverted toward baseline during the 2-h period following discontinuation of the infusion. We conclude that the hypoxemia occurring in the early phase of group B streptococcal sepsis does not develop solely because of increased shunt, but rather is produced by a decline in cardiac output in conjunction with mismatching of pulmonary perfusion to alveolar ventilation.

Nitric oxide (NO), an important vasodilatory modulator of systemic and pulmonary vascular tone, is synthesized from l-arginine by the enzyme NO synthase in vascular endothelial and smooth muscle cells. l-Arginine analogs, such as Nω-nitro-l-arginine methyl ester (l-NAME), are competitive antagonists of NO synthase and inhibit NO synthesis. Group B streptococcus (GBS) causes pulmonary hypertension, hypoxemia, lung vascular injury, and reduced cardiac output in both human newborns and neonatal piglets. Lung vascular injury associated with prolonged GBS infusion in piglets may attenuate NO production and thus promote severe pulmonary hypertension. We studied the effect of the NOS inhibitor, l-NAME and the precursor of NO, l-arginine, on pulmonary and systemic hemodynamics during late-phase GBS sepsis in the piglet model. Neonatal piglets were anesthetized, ventilated with room air, and randomized to receive a continuous infusion of saline (n5) or GBS (n5) for 4 h. After 3 h of infusion, both groups received a bolus of l-NAME (3 mg/kg). Hemodynamic and gas exchange indices were measured at baseline, 30 min, and 3 h of infusion, and 30 min and 1 h after l-NAME treatment. l-NAME treatment caused 1) significant increases in mean pulmonary arterial pressure, pulmonary vascular resistance, mean systemic arterial pressure, and systemic vascular resistance for both groups; 2) a similar percentage of increase in pulmonary vascular resistance for the two groups; 3) greater reduction in cardiac output and SV in the GBS compared with the control group; and 4) no significant alterations in arterial partial pressure of oxygen or the difference between alveolar and arterial partial pressure of oxygen for either group. l-Arginine Arginine (1 g/kg) infusion after 3 h of GBS infusion (n3) caused no significant changes in any measured hemodynamic or gas-exchange variable. We conclude that 1) endogenous NO synthesis is ongoing during late-phase GBS-induced pulmonary hypertension in neonatal piglets, and 2) NO synthesis is not limited by the substrate l-arginine in this model. NO synthase inhibitors alone appear to be contraindicated in the treatment of neonatal GBS sepsis due to worsening pulmonary hypertension and progressive decline in cardiac output.

The endogenously produced cytokine, tumor necrosis factor-α (TNF-α), has been shown in adult animal models to be associated with many of the pathophysiologic effects of sepsis, including systemic hypotension and hemorrhagic necrosis. TNF-α can induce the release of various vasoactive arachidonic acid metabolites, suggesting that TNF-α may act either directly or via intermediary substances in producing its effects. The pathophysiologic role of TNF-α in neonatal sepsis, especially its potential effect on pulmonary vascular tone, is presently unknown. To assess the role of TNF-α in neonatal sepsis, 19 piglets (19 ± 5 d old) were anesthetized, intubated, paralyzed, mechanically ventilated, and catheterized to assess pulmonary and systemic vascular hemodynamics and pulmonary gas exchange. The multiple inert gas elimination technique was used to assess ventilation perfusion matching. A 30-min infusion of human recombinant TNF-α (250 μg/kg total dose) was administered to animals pretreated with either 10 mg/kg dazmegrel, a thromboxane synthase inhibitor (n9) or placebo (n10). TNF-α alone induced a prompt and sustained rise in pulmonary arterial pressure and pulmonary vascular resistance that continued at least for 2 h after onset of the infusion. In contrast, the animals pretreated with dazmegrel demonstrated no rise in pulmonary vascular resistance until 2 h after the onset of the infusion. Neither group of animals demonstrated a significant decline in arterial PaCO2or evidence from inert gas analysis of A/ mismatching or increase in intrapulmonary shunt. We conclude that human recombinant TNF-α induces an acute and sustained elevation in Ppa and PVR without the concomitant development of hypoxemia or intrapulmonary shunt. The pulmonary hypertension is blocked acutely by the putative TxA2synthase inhibitor, dazmegrel.

Group B ß-hemolytic streptococcus (GBS) causes thromboxane (Tx)-associated pulmonary hypertension and hypoxemia in neonatal animals and human infants. The components of GBS that induce these features of sepsis are incompletely characterized. The capsular polysaccharide has been implicated based on the effects of GBS extracts. We used isogenic mutants of a parent GBS strain (COH 31 rs) devoid of capsular polysaccharide or ß- hemolysin to determine if these components caused the acute features of GBS bacteremia. In neonatal piglets, we observed a similar increase in pulmonary vascular resistance (PVR, mm HgLmin) during a 1 h infusion at 5 x 108colony-forming unitkgh of COH 31 rs (n5, 11.6 ± 1.4 to 67.1 ± 17.9), an isogenic GBS mutant devoid of type III CP (n5, 12.5 ± 1.4 to 56.9 ± 5.0), and an isogenic GBS mutant devoid of ß-hemolysin (n4, 11.0 ± 1.9 to 51.9 ± 7.9). All three GBS strains caused increases in blood TxB2levels, mild arterial hypoxemia, mild reduction in mixed venous PO2and a 30-40 reduction in cardiac output after a 1 h infusion. The Tx-synthase inhibitor, dazmegrel, completely reversed pulmonary hypertension, and partially reversed arterial hypoxemia and TxB2levels to baseline values for all GBS strains. In six additional piglets, infusion of polystyrene beads of similar size to GBS at a dose of 5 x 108beadskgh caused no changes in gas exchange or blood TxB2levels, but a mild increase in PVR (13.3 ± 2.0 to 17.7 ± 3.5). This suggests a nonspecific response to circulating particulates is not the major cause of the acute features of GBS bacteremia in piglets. We conclude that type III capsular polysaccharide and ß-hemolysin are not essential for type III GBS to cause acute Tx-associated pulmonary hypertension and hypoxemia in piglets.

Background:Although inhaled nitric oxide (iNO) causes selective pulmonary vasodilation and improves oxygenation in newborn infants with persistent pulmonary hypertension, its effects are variable. We hypothesized (1) that the response to iNO therapy is dependent on the primary disease associated with persistent pulmonary hypertension of the newborn (PPHN) and (2) that the combination of high-frequency oscillatory ventilation (HFOV) with iNO would be efficacious in patients for whom either therapy alone had failed.

• A review of all infants admitted to the two intensive care nurseries in Seattle from July 1, 1980, through Dec 31, 1981, was performed to evaluate the outcome of infants still requiring supplemental oxygen and/or mechanical ventilation at 1 month of age. Sixty-three Infants were identified. Fifty-six infants survived to at least 2 years of age, including 11 of 13 in the subgroup of infants requiring 40% or more oxygen at 1 month of age. Eight (14%) of the 56 survivors have required prolonged rehospitalization for pneumonia or other respiratory illnesses in the first two years following birth. We conclude that the degree of gas exchange impairment assessed at 1 month of age does not predict ultimate outcome from neonatal chronic lung disease.(AJDC 1985;139:351-354)

We used an immunoperoxidase method to examine the expression of the immune activation marker HLA-DR on pulmonary tissue obtained at autopsy from 14 patients dying of bronchopulmonary dysplasia. Controls consisted of 16 age-matched, sex-matched children dying of noncardiac, nonrespiratory, noninfectious illnesses or as a result of motor vehicle accidents. We did not observe aberrant expression of HLA-DR on pulmonary cndothelial cells. Positive staining appeared exclusively on macrophages. We quantitated the expression of antigen by counting the number of positive macrophages and total macrophages/monocytes per high power field. Bronchopulmonary dysplasia patients displayed significantly greater numbers of both positive and total macrophages compared with the control group (p< 0.05). The percent positive macrophages also was significantly higher in the bronchopulmonary dysplasia patients (p< 0.005). We also examined a group of patients dying with infant respiratory distress syndrome. There was no significant difference in number of total macrophages in this group compared with age-matched controls.

The methylxanthine derivative pentoxifylline (PTF) demonstrates vasodilatory properties in vivo.We tested the hypothesis that PTF infusion would blunt or inhibit tumor necrosis factor-α (TNFα)-induced and U46, 619-induced increases in mean pulmonary artery pressure and pulmonary vascular resistance (PVR) in the neonatal piglet and would do so by altering production of eicosanoid vasoactive mediators. Anesthetized, paralyzed piglets (age 10–29 d) were randomized and treated with a 30-min infusion of TNFαalone (n= 13 animals), with a combination of TNFαplus pretreatment and continuous infusion with PTF (n= 6), or with a combination of U46, 619 for 30 min plus pretreatment and continuous infusion of PTF (n= 5). There was no difference in pulmonary or systemic hemodynamic indices between the three groups at baseline. PVR was significantly elevated at 15 min and at 2 h in the TNFα-only group. The TNFα-induced rise in mean pulmonary artery pressure and PVR was inhibited by the PTF until 2 h, by which time PVR was elevated above baseline and was comparable to the value found in animals treated with only TNFα. PTF produced no inhibition in the U46, 619-induced elevation of PVR during the 30-min simultaneous treatment. In the PTF + TNFαgroup, mean systemic blood pressure declined to 50 of baseline value (p< 0.02) by 2 h of age. No significant decline was noted in mean systemic arterial pressure of the TNFα-only or the U46, 619-treated group. Plasma 6-keto-prostaglandin F1αincreased above baseline values by 2 h only in the PTF + TNFα-treated group (p< 0.02); no significant change from baseline in thromboxane B2levels was found in any experimental group. We conclude that treatment with PTF transiently inhibited the TNFα-induced elevation in mean pulmonary artery pressure and PVR but was associated also with significant systemic hypotension. PTF failed to counteract elevated PVR produced by the thromboxane mimetic, U46, 619.

The hemodynamic and renal effects of atriopeptin II were investigated in rats, using electromagnetic flowmeters and thermodilution and clearance techniques, and its direct cardiac effects were studied in isolated rat atria and ventricular muscle strips. Atriopeptin II had no effect on the rate or force of contraction of rat cardiac tissue in vitro. In anesthetized rats, i.v. injections of 1, 3, 5, and 7 mUg/kg induced only transient hemodynamic effects: blood pressure (BP) was briefly reduced, and renal blood flow (RBF) but not mesenteric blood flow (MBF) increased. Infusions over 30 min at rates of 0.3, 1, 3, and 10 mUg/kg/min caused a fall in BP, mediated by a reduction of cardiac output (CO); RBF and MBF were decreased in a dose-related manner, and systemic and regional vascular resistances rose. A reduction of RBF was also observed in clearance experiments, but glomerular filtration rate remained unchanged and the filtration fraction increased significantly. Natriuresis occurred at all rates of atriopeptin tested (0.3, I, and 3 mU-g/kg/min i.v.) These results suggest that i.v. injection of atriopeptin II induces transient hypotensive and regional vasodilatory effects. Upon i.v. infusion, BP is lowered by way of a reduction in CO, which is accompanied by systemic and regional vasoconstriction. The decrease in CO cannot be ascribed to a direct cardiodepressant action, and the effects on regional blood flows are probably due to reflex activation. The natriuretic effects of atriopeptin II are independent of renal vasodilatation and may be attributable to changes in intrarenal hemodynamics or to direct tubular effects.

ABSTRACT: We used an immunoperoxidase method to examine the expression of the immune activation marker HLA-DR on pulmonary tissue obtained at autopsy from 14 patients dying of bronchopulmonary dysplasia. Controls consisted of 16 age-matched, sex-matched children dying of noncardiac, nonrespiratory, noninfectious illnesses or as a result of motor vehicle accidents. We did not observe aberrant expression of HLA-DR on pulmonary cndothelial cells. Positive staining appeared exclusively on macrophages. We quantitated the expression of antigen by counting the number of positive macrophages and total macrophages/monocytes per high power field. Bronchopulmonary dysplasia patients displayed significantly greater numbers of both positive and total macrophages compared with the control group (p < 0.05). The percent positive macrophages also was significantly higher in the bronchopulmonary dysplasia patients (p < 0.005). We also examined a group of patients dying with infant respiratory distress syndrome. There was no significant difference in number of total macrophages in this group compared with age-matched controls.

The methylxanthine derivative pentoxifylline (PTF) demonstrates vasodilatory properties in vivo. We tested the hypothesis that PTF infusion would blunt or inhibit tumor necrosis factor-a (TNFa)-induced and U46, 619-induced increases in mean pulmonary artery pressure and pulmonary vascular resistance (PVR) in the neonatal piglet and would do so by altering production of eicosanoid vasoactive mediators. Anesthetized, paralyzed piglets (age 10–29 d) were randomized and treated with a 30-min infusion of TNFaalone (n = 13 animals), with a combination of TNFaplus pretreatment and continuous infusion with PTF (n = 6), or with a combination of U46, 619 for 30 min plus pretreatment and continuous infusion of PTF (n = 5). There was no difference in pulmonary or systemic hemodynamic indices between the three groups at baseline. PVR was significantly elevated at 15 min and at 2 h in the TNFa-only group. The TNFa-induced rise in mean pulmonary artery pressure and PVR was inhibited by the PTF until 2 h, by which time PVR was elevated above baseline and was comparable to the value found in animals treated with only TNFa. PTF produced no inhibition in the U46, 619-induced elevation of PVR during the 30-min simultaneous treatment. In the PTF + TNFagroup, mean systemic blood pressure declined to 50% of baseline value (p < 0.02) by 2 h of age. No significant decline was noted in mean systemic arterial pressure of the TNFa-only or the U46, 619-treated group. Plasma 6-keto-prostaglandin F1aincreased above baseline values by 2 h only in the PTF + TNFa-treated group (p < 0.02); no significant change from baseline in thromboxane B2levels was found in any experimental group. We conclude that treatment with PTF transiently inhibited the TNFa-induced elevation in mean pulmonary artery pressure and PVR but was associated also with significant systemic hypotension. PTF failed to counteract elevated PVR produced by the thromboxane mimetic, U46, 619.

ABSTRACT: Thromboxane-associated pulmonary hypertension occurs in animals during intravenous infusion of group B streptococcus (GBS), a gram-positive neonatal pathogen. We postulated that other gram-positive neonatal pathogens, such as Streptococcus fecalis (ENT) and Staphylococcus epidermidis (S. epi) would also induce increased thromboxane synthesis and pulmonary hypertension when infused into piglets. We observed similar hemodynamic and gas exchange abnormalities during stepwise increases in the dose of GBS, Ent, and S. epi (n = 3, 4, and 4 piglets receiving each bacteria, respectively). Pulmonary vascular resistance increased significantly in the absence of acidosis or reduced arterial or mixed venous pO2at a dose of 2.5 × 108cfu/kg/h for Ent and S. epi. In 14 additional piglets, pulmonary vascular resistance increased markedly after 60 min of intravenous infusion of 4 ± 1 × 108cfu/kg/h for each organism (p < 0.05, GBS: 11.7 ± 1.8 to 75.6 ± 18.4 mm Hg/liter/min, Ent: 12.7 ± 1.7 to 64.9 ± 10.6 mm Hg/liter/min, S. epi: 10.5 ± 0.8 to 56.9 ± 6.0 mm Hg/liter/min), and blood thromboxane B2 levels increased (p < 0.05, GBS: 30 ± 10 to 1830 ± 330 pg/nl, Ent: 20 ± 7 to 1110 ± 300 pg/ml, S. epi: 31 ± 9 to 1260 ± 350 pg/ml). This dose of each bacteria caused a similar degree of mild arterial hypoxemia (57–66 mm Hg). The thromboxane synthetase inhibitor, dazmegrel, completely reversed pulmonary hypertension, reduced TxB2 levels to near baseline values, and partially reversed arterial hypoxemia despite ongoing bacterial infusion. We conclude that thromboxane-associated pulmonary hypertension occurs in piglets during infusion of different gram-positive neonatal pathogens.

Objective: To quantitate airway muscle changes in infants born at 23 to 41 weeks' gestation (control subjects) and to compare the changes with those in infants with chronic lung disease. Methods: Fifty-five human lungs (from infants born at 23 to 41 weeks' gestation) were studied: 46 from infants who died of various diseases within 72 hours of birth, and 9 from infants with CLD (infants born at 26.9 +/- 0.5 weeks' gestation, who lived 17 +/- 8 days). All the lungs were perfused via the trachea and pulmonary artery in a standardized protocol. Formalin-fixed tissues in paraffin blocks were cut 5 @mm thick. Sections were immunohistochemically stained for @a-smooth muscle actin. By using computerized image analysis to quantitate images digitized into the computer, we measured the area of muscle, epithelium, airway lumen, and length of basement membrane in 18 airways, from the smallest bronchioles to bronchi, in each infant. Results: Muscle was present at 23 weeks' gestation at all levels of the bronchial tree, and from 25 weeks to term the control lungs had a similar quantity of muscle at any given airway circumference. Relative to airway size, there was more muscle in small airways, less than 1000 @mm in circumference, than in larger airways. In airways greater than 1500 @mm in circumference, infants with CLD had significantly more muscle than did control lungs. Conclusions: Airway muscle is present at 23 weeks' gestation at all levels of the conducting airways. The 25-week gestation infants had a quantity of airway muscle relative to airway circumference similar to that of term infants. Preterm infants with CLD who were aged 9 to 29 days have increased airway muscle in airways greater than 1500 @mm in circumference. Bronchospasm in very low birth weight infants is possible within the first days of life. (J Pediatr 1997;130:570-6)

This study examined the hemodynamic and regional vascular effects of orally administered nisoldipine, a new dihydropyridine derivative (0.3 mg/kg) in normal conscious rats (n 10). Nisoldipine significantly reduced systemic vascular resistance (0.58 to 0.38 mm Hg kg min/ml, p < 0.05) and mean arterial pressure (122 to 108 mm Hg, p < 0.05), and increased heart rate (395 to 447 beats/min, p < 0.01) and cardiac index (225 to 326 ml/beat/kg, p < 0.05). Left ventricular end-diastolic pressure was slightly decreased by nisoldipine (9.6 to 3.8 mm Hg, p < 0.05). Blood flow (radioactive microspheres, 15 ± 5 μm in diameter) to heart, gut, kidney, and brain was significantly increased. Improvement of blood flow was most pronounced in the coronary circulation (58) followed by the gut and renal circulatory beds. We conclude that nisoldipine represents a new orally effective calcium antagonist with highly selective effects in vascular smooth muscle as compared with its direct cardiac effects. The results are compared with our previous study utilizing intravenous nisoldipine.

ABSTRACT. The appearance of phosphatidylglycerol in the tracheal wash of infants with hyaline membrane disease (HMD) has been reported to be associated with clinical signs of recovery. We analyzed lung tissue and bronchoalveolar lavage surfactant in an animal model of HMD to determine whether phosphatidylglycerol or some other component is necessary for recovery. The amount and composition of phospholipid (PL) was determined in the premature Macaca nemestrina monkey (140 days' gestation) during an acute stage of HMD, and in two stages of recovery. These changes were compared to observations made in healthy premature controls (140 days), gestational age-matched fetuses (140 days), and fetuses of 150 days' gestation (term=168 days). The amount of PL and its surfactant composition in lung homogenates of the right lower lobe and in lavage of the excised left lung was determined. Compared to 140-day fetuses, the healthy controls had a several-fold increase in lavage PL and disaturated phosphatidylcholine (DSPC) during the first few days of life (p<0.05). Prior to recovery, animals with HMD had no such increase in lavage PL or DSPC and demonstrated poor deflation stability. Recovery was associated with increased tissue and lavage PL (p<0.05) and increased fractions of phosphatidylinositol and DSPC (p<0.05), but not phosphatidylglycerol. The tissue compositional changes observed during recovery reflected maturational changes observed in the fetal animals studied at 10 days' greater gestational age. Increases in lavage DSPC beyond 2 mg/g dry lung were not associated with further increases in deflation stability. We conclude that increases in DSPC and phosphatidylinositol are associated with recovery from HMD but phosphatidylglycerol does not appear necessary for either improved deflation stability or clinical recovery.

Hyaline membrane disease (HMD), the most common life-threatening respiratory disorder of newborns, is associated with lung injury manifested by alveolar proteinaceous edema. The cause of the disease is thought to be elevated alveolar surface tension due to surfactant deficiency at birth. Treatment with exogenous surfactant may be unsuccessful due to problems in distribution of the surfactant, or inhibition of the surfactant by alveolar proteinaceous edema. Liquid ventilation with oxygen-saturated perfluorocarbon liquid has been proposed as a method to eliminate alveolar surface tension; little is known about the interfacial tension between perfluorocarbon liquids and the lung lining layer. Premature and term newborn monkeys were treated from birth with a pressure-limited, time-cycled liquid ventilator using oxygenated perfluorocarbon liquids (APF-145 and perflubron). Adequate gas exchange was achieved, and pilot experiments suggest long-term survival without adverse sequelae. Although many questions remain, liquid ventilation is a promising tool for the prevention and treatment of lung injury in newborns.

Non-cyclooxygenase-derived prostaglandin F2- (PGF2)-like compounds can be formed by membrane lipid peroxidation. We sought to characterize one member of this class of compounds, 8-epi-PGF2α, for its biological properties on the pulmonary vasculature in young piglets. We first compared 8-epi-PGF2α to a thromboxane (Tx) mimetic, U46,619, to determine relative pulmonary vasoconstrictive effects. We next determined if the vasoconstriction induced by both agonists would be reversed by 50 ppm of inhaled nitric oxide (NO). We also determined the degree of inhibition of U46,619 and of 8-epi-PGF2α by a Tx receptor antagonist, SQ 30,741. Anesthetized, ventilated piglets (18 ± 2 days, 3.7 ± 0.5 kg) were infused with randomly selected doses of U46,619 and of 8-epi-PGF2α to describe dose-response curves plotting pulmonary vascular resistance (PVRi) against dose. When a maximal dose was achieved, inhaled NO (50 ppm) was administered. Results for the calculated PVRi (mm Hg/l/min/kg) are as follows (mean ± SD): for U46,619, baseline value: 43 ± 10, peak dose: 165 ± 46, and peak dose with NO: 73 ± 33; for 8-epi-PGF2α, baseline value: 38 ± 9, peak dose: 138 ± 30, and peak dose with NO: 72 ± 25, and for both drugs, PVRi at peak dose was greater than at baseline (p < 0.001). PVRi at peak dose plus NO also remained elevated above baseline (p < 0.05; repeated measures ANOVA). The Tx receptor antagonist SQ 30,741 (1–10 mg/kg) inhibited U46,619-induced pulmonary vasoconstriction completely; however, the 1 mg/kg dose provided only 90% inhibition against 8-epi-PGF2α. We conclude that 8-epi-PGF2α can contribute to pulmonary vasoconstriction in young piglets in a dose-dependent reversible manner and that it acts primarily via pulmonary vascular Tx receptors.

Summary: Cholestasis develops in many infants supported with extracorporeal membrane oxygenation. We prospectively investigated the role of hemolysis and di-(2-ethylhexyl) phthalate exposure in the development of this cholestasis. Both di-(2-ethylhexyl) phthalate levels and hemolysis, as measured by maximum free hemoglobin, were significantly (p < 0.025) associated with the degree of cholestasis. Other clinical and laboratory factors that may contribute to cholestasis were also investigated and not found to be related to the degree of cholestasis. We speculate that hemolysis during extracorporeal membrane oxygenation support produces a large bilirubin load whose excretion is inhibited by mechanisms similar to the inspissated bile syndrome and/or by di-(2-ethylhexyl) phthalate. This would result in a predominantly direct hyperbilirubinemia with little evidence of hepatocellular or canalicular injury.

We studied the effects of dopamine HCI, a potent vasoactive catecholamine, on ventilation-perfusion matching and pulmonary hemodynamics in infant lambs breathing room air and a hypoxic gas mixture. Neither alveolar hypoxia alone, nor dopamine infusion alone during room air breathing, significantly altered the ventilation-perfusion (VA/Q) distribution or intrapulmonary shunt size. Dopamine infusion during room air breathing did increase pulmonary blood flow (PBF; p < 0.05). Dopamine infusion plus alveolar hypoxia was associated with a significant increase in minute ventilation, PBF, and pulmonary arterial pressure (Ppa), a doubling (p < 0.01) of the fraction of PBF to low VA/Q lung regions, and a significant increase (p < 0.05) in intrapulmonary shunt when compared to results obtained during hypoxic gas breathing without dopamine infusion. We conclude the dopamine infusion increased PBF during room air breathing, and perturbed ventilation-perfusion matching and increased Ppa and minute ventilation with no net change in arterial pO2 during the breathing of a hypoxic gas mixture.

The effects of i.v. captopril on regional blood flow (radioactive microspheres, 15 +/- 5 micron), hemodynamics and maximal oxygen consumption were evaluated in conscious rats with congestive heart failure due to large myocardial infarction (n = 9, infarct size 39.5 +/- 2% of left ventricle) and compared to data obtained from rats subjected to sham surgical procedures (n = 8). In both groups data were obtained at rest and during submaximal treadmill exercise during alternate infusion of captopril and saline. In the congestive heart failure group captopril reduced systemic vascular resistance, mean arterial pressure and left ventricular systolic pressure (P less than .05 each). Blood flow to the renal, gastrointestinal and coronary circulations was reduced in the heart failure group treated with saline vehicle. Flow to the renal and gastrointestinal beds of heart failure animals was enhanced to values similar to those observed in sham animals during captopril treatment. Left ventricular coronary flow was also increased significantly by captopril in both sham and heart failure animals. The most prominent effects of captopril occurred in the renal circulation of the heart failure group in which blood flow increased by 55%. Blood flow to skeletal muscle and skin was unchanged by captopril both in sham and heart failure animals at rest and during exercise. Maximal oxygen consumption was not affected by captopril treatment. Thus, captopril induced a differential pattern of vasodilation with the greatest effect in the renal bed and a less intensive effect in the gastrointestinal and coronary beds. The unchanged flow to skeletal muscle may explain the failure of captopril to improve exercise capacity after short-term administration.

ABSTRACT: Acute hypoxic pulmonary vasoconstriction is attenuated by respiratory alkalosis. It is unknown if alkalosis similarly reduces pulmonary vasoconstriction produced by thromboxane A2. Respiratory alkalosis does not always attenuate persistent pulmonary hypertension in newborns, some of whom have elevated serum thromboxane B2 levels. We hypothesized that alkalosis attenuates thromboxane-induced pulmonary vasoconstriction less than it does hypoxic pulmonary vasoconstriction in infants. Hemodynamic responses to respiratory alkalosis during pulmonary vasoconstriction produced in random order by breathing 12% inspired oxygen and by infusing 0.1 µg/kg/min of the thromboxane-mimetic U46,619 were compared in eight 2-wk-old piglets. Hypoxia increased mean pulmonary artery pressure from 12 ± 3 to 29 ± 2 mm Hg and pulmonary vascular resistance (PVR) from 11 ± 4 to 25 ± 8 mmHg/L/min; U46,619 increased pulmonary artery pressure from 16 ± 5 to 37 ± 6 mm Hg and PVR from 14 ± 5 to 51 ± 17 mm Hg/liter/min. U46,619 also decreased cardiac output accounting in part for the greater increase in PVR compared to hypoxia-induced vasoconstriction. Respiratory alkalosis decreased PVR to 14 ± 6 mm Hg/liter/min during exposure to hypoxia and to 28 ± 9 mm Hg/liter/min during infusion of U46,619. In six additional piglets with U46,619-induced pulmonary vasoconstriction, the effects of lung stretch and hypocapnic alkalosis were separated by doubling tidal volume and then adding inspired CO2to return PaCO2to prehyperventilation levels. Respiratory alkalosis decreased PVR from 52 ± 36 to 35 ± 21 mm Hg/liter/min. Despite the increased tidal volume, PVR increased to 53 ± 35 Hg/liter/min when PaCO2returned to 44 ± 5 mm Hg. Respiratory alkalosis rather than lung stretch reduces thromboxane-induced and hypoxia-induced pulmonary vasoconstriction by equal proportions.

Reading comprehension was studied as a function of individual differences in extraversion-introversion, neuroticism and academic and resultant achievement motivation. Seventy-eight college students were categorized into personality or motive groups on the basis of personality test scores, dividing the distribution into top, middle and bottom thirds. Academic achievement motivation was assessed by a recently developed measure previously used with British students; resultant achievement motivation was measured as need for achievement minus fear of failure. Analyses of variance indicated no significant contributions of any of the IDs studied to reading comprehension. Discussion centered on limitations of the study, its relationship to previous work, and future directions.

Investigations over a 5‐year period on deep plowing, subsoiling, and deep incorporation of lime and fertilizer were conducted on Almena silt loam (strongly acid and tight subsoil). Subsoil treatments included tillage only, tillage with liming, and tillage with liming and fertilizing. The plow‐layer of all subsoiled plots was limed to pH 6.5 and fertilized uniformly. Plots not subsoiled received varying amounts of fertilizer, equal in several cases to the total applied to both the plow‐layer and subsoil of the subsoiled plots. Deeper root penetration of alfalfa was promoted by subsoil liming and fertilizing, but not by subsoiling alone. Subsoil liming and fertilizing helped materially in establishing alfalfa during a dry year. Second‐year and especially third‐ and fourth‐year hay yields appeared to be increased (up to one‐half ton per acre in dry years) by subsoil liming and fertilizing, but were not increased by subsoil treatment where the fertilizer was omitted. The potassium content of the alfalfa from subsoil fertilized plots was as much as 1% higher during dry years. Corn and oats did not appear to respond to these subsoil treatments even in the dry years, and when they followed the alfalfa‐brome grown with subsoil treatments. In the case of deep plowing with liming and fertilizing, first‐year alfalfa‐brome gave a consistent but small response to this deep treatment; however, second‐year hay yields were increased appreciably when plowing was to a 12‐inch depth. Where the depth of plowing was 9 inches, the response was somewhat less, but still greater than where plowing, liming, and fertilizing was to a 6‐inch depth only. In no year were the yields of corn and oats increased by deeper plowing, liming, and fertilizing, although increased vigor was noted in some years.

The influence of the pH of the culture medium on the availability of rock phosphate to oats was investigated by means of two sets of pot cultures. In one set, quartz sand containing 1% of montmorillonite clay served as the culture medium. Samples of the clay, previously saturated with H+, were treated with varying amounts of Ca(OH)2to provide a series of samples having pH values of 4.9, 5.5, 6.2, 6.7, and 7.4, respectively. Addition of these clays to quartz cultures imparted their respective pH's to the cultures. In one series, rock phosphate served as a source of phosphorus, and in a second, superphosphate served similarly. All other nutrients were supplied in soluble forms. The yield with rock phosphate was highest at pH 5.5. With superphosphate yields were good throughout and approximately double those with rock phosphate. A soil pH of 6.0 or lower appears to be necessary for satisfactory utilization of rock phosphate. In a second set of pot cultures, Spencer silt loam subsoil of pH 5.1 and very low in organic matter and available phosphorus served as the culture medium. Other nutrients than phosphorus were supplied in available forms. When rock phosphate was applied 1 month prior to liming to pH 7, the oats grew much better and absorbed much more phosphorus than when the order of applications was reversed, showing that strongly acid soils act on rock phosphate rather rapidly, and that for rock phosphate to be effective acidity is needed. Thus, it is advisable to apply rock phosphate a year or more in advance of liming.

The factors involved in fixation of phosphorus, added as Ca(H2PO4)2· H2O, by montmorillonite clay were investigated. The influence of the free iron and aluminum oxides, which accompany this clay, on the fixation was determined by measuring fixation before and after removal of these oxides. It was found that the free iron and aluminum oxides accounted for most of the fixation against weak acid extraction. No appreciable fixation against water extraction occurred when the free iron and aluminum oxides were removed and the clay was H‐saturated. The influence of saturation of the clay with the exchangeable cations usually found in soils on phosphorus fixation was investigated. Ca‐saturation clay, treated to remove free iron and aluminum oxides, fixed large amounts of phosphorus against water extraction. However, this phosphorus was recovered by weak acid extraction. The evidence obtained in this investigation indicates that the phosphorus fixed by Ca‐saturated clay is fixed as calcium phosphate complexes, and not by a H2PO4‐Ca‐clay bonding as has been proposed. Clays which were saturated with Na, K or Mg, fixed only small amounts of phosphorus against water extraction.

A study was made of the effect on soil aggregation of microbial gums produced in vitroand in soil. Attention was also given to the effects of fertility treatments, kind of crop, and season on the aggregation of Spencer silt loam. The bacterial polysaccharide synthesized by Agrobacterium radiobacterin pure culture had a marked aggregating effect when added to Spencer and Miami silt loams. Addition of as little as 0.02 gm. of this gum to 100 gm. of soil caused a 50% increase in aggregates > 0.1 mm. in diameter. The gum content of numerous samples of Miami and Spencer silt loams correlated well with their respective levels of aggregation. The addition of extracted soil gum to a poorly aggregated soil increased significantly the degree of aggregation. Undecomposed pulverized plant residues added to soil increased the gum content to a maximum in six days. The maximum percentage of aggregates > 0.5 mm. in diameter was not reached until the twelfth day. In a four‐year rotation of corn, oats, and two years of alfalfa‐brome on plots which were variously limed and fertilized, increases in percentage of soil aggregates occurred at pH 6.0, 6.5, and 7.5 as compared with the aggregation level of the check plot (pH 5.5). High levels of available phosphorus and potassium as compared with low levels promoted aggregation. In general, structural deterioration began under corn and continued until the following crop of oats attained considerable growth. Thereafter improvement continued and reached a maximum during the second year of hay. Seasonal variations in soil aggregation occurred on all plots, irrespective of soil treatment or crop, and was characterized by a gradual increase in percentage of water‐stable aggregates during the spring and summer followed by a sharp decline in September.

After making preliminary tests of methods proposed by other investigators for determining the fixed potassium of soils and also of several new approaches, the approach found most promising — heating at 500° C for two hours — was investigated in considerable detail. It was found that on heating ammonium saturated soils, illite, and bentonite in this manner, substantial release of nonexchangeable or fixed potassium could be effected without breakdown of primary potassium minerals. Not all of the fixed potassium can be released in one heating, the amount thus released representing an equilibrium point related to the amount which still remains fixed. The procedure finally adopted consists in heating ammonium saturated soil at 500° C for two hours, then extraction of the potassium made exchangeable with ammonium acetate solution, and finally determination of the potassium in the solution by means of the flame photometer. This heating, extraction, and determination are repeated once, and from the amount of fixed potassium released during the second heating, the total amount of fixed potassium present may be calculated by use of a summation formula for a geometric progression. The method is simple and is adaptable to rapid analysis of large numbers of samples. On the basis of the results obtained with soils and minerals, it is believed that the method is reasonably specific for fixed potassium, and offers the possibility of being a practicable and reliable method.