Search

Objective: The purpose of this study was to evaluate triple-phase helical CT for detection of hepatic metastases from breast carcinoma., Subjects and Methods: Breast cancer patients were studied prospectively with triple-phase helical CT in 300 consecutive examinations. Hepatic arterial-dominant and portal venous-dominant phase scans were initiated at 20 and 65 sec, respectively, after IV injection of 175 ml of iopamidol (30 mg/ml) at 5 ml/sec. Three independent observers each reviewed 200 cases of the portal venous-dominant phase for lesion number, conspicuity, and attenuation. Subsequently, portal venous-dominant phase images were reevaluated in conjunction with hepatic arterial-dominant phase or unenhanced images., Results: Hepatic metastases were identified in 79 (26%) of 300 cases. Lesions detected on portal venous-dominant, hepatic arterial-dominant, and unenhanced images were as follows: observer 1, n = 198, 164, and 171; observer 2, n = 254, 233, and 233; and observer 3, n = 291, 270, and 276 (p > .05). The mean total lesion count was 387, with more lesions detected on portal venous-dominant phase than on either hepatic arterial-dominant phase or unenhanced images (p < .001 and p < .0001, respectively). For individual observers, 10-26% of lesions were hypervascular on hepatic arterial-dominant phase images. Two to 4% of lesions were identified only on hepatic arterial-dominant phase or unenhanced images. However, in these few cases, the lesions either were false-positives or were seen in conjunction with additional metastases on portal venous-dominant images., Conclusion: Routine use of triple-phase CT in patients with breast carcinoma may not be warranted: Addition of the hepatic arterial-dominant phase or unenhanced images revealed few additional lesions in our group of 300 patients.

Objective: The purpose of this study is to determine the safety and effectiveness of percutaneous imaging-guided biopsy in the diagnosis of focal splenic lesions., Materials and Methods: From May 1995 to November 1997, 20 imaging-guided biopsies of focal splenic lesions were performed in 18 patients, including seven patients with a prior diagnosis of extrasplenic malignancy (breast cancer, n = 3; lymphoma, n = 2; ovarian cancer, n = 1; and osteogenic sarcoma, n = 1), three immunosuppressed patients (cause of immunosuppression: AIDS, n = 1; liver transplantation, n = 1; and bone marrow transplantation, n = 1), two patients with anemia, one patient with a recent history of IV drug abuse, and five patients with incidentally discovered splenic lesions. Biopsies were performed with an 18-gauge (n = 1), a 20-gauge (n = 8), or a 22-gauge (n = 14) self-aspirating needle or an 18-gauge cutting needle (n = 1). Biopsies were considered successful if a specific diagnosis of benign or malignant disease was made., Results: A specific diagnosis was made in 16 (88.9%) of 18 patients, and no complications occurred. Malignancy was diagnosed in six patients including three patients with lymphoma. Benign conditions were diagnosed in 10 patients: a cyst in two patients; hamartoma in one; lipogranuloma in one; infarct in one; and infection in four, including one case each of Candida albicans, Pneumocystis carinii, Mycobacterium tuberculosis, and mixed flora. The tenth benign diagnosis was a pseudotumor of the spleen related to a bulbous tail of the pancreas that was inseparable from the splenic hilum. Biopsy did not establish a diagnosis in one patient with lymphoma and in one patient with presumed splenic candidiasis. A mean of 1.5 needle passes was made per biopsy., Conclusion: Imaging-guided splenic biopsy is a safe technique that provides a specific diagnosis in most patients with focal splenic lesions.

Purpose: To review the normal radiologic appearance of pancreatic transplants that use portal venous and enteric drainage, and to review the appearance of a variety of postoperative complications., Method: We retrospectively reviewed the computed tomographic (CT) scans, magnetic resonance (MR) images, and ultrasounds of patients who had undergone pancreatic transplantation using portal venous and enteric drainage., Results: At CT, the normal pancreatic transplant appears as a heterogeneous mass composed of pancreatic parenchyma, vessels, and omental wrap. On MR imaging, a normal transplant demonstrates intermediate signal intensity on T1- and T2-weighted sequences. Sonographic evaluation of a normal transplant reveals a hypoechoic gland that contains readily detectable low-resistance arterial and venous Doppler waveforms. Acute postoperative complications include acute rejection, which has a nonspecific radiologic appearance, and transplant pancreatitis, which is often manifested on CT by stranding of the peritransplant fat. Chronic postoperative complications include small bowel obstructions, graft pancreatitis secondary to obstruction of the Roux loop, and chronic rejection., Conclusion: Knowledge of the radiologic appearance of the normal pancreatic transplant is required before transplant-related complications can be detected.

Background: Improved resectability is a major theoretical benefit of preoperative chemoradiation for pancreatic cancer. Since 1994, patients at Duke University Medical Center with locally advanced pancreatic cancer have been treated with multimodality preoperative therapy. The purpose of this study was to review our experience with preoperative therapy for locally advanced pancreatic cancer and determine if an aggressive neoadjuvant regimen would not only downstage these tumors pathologically but also improve the odds of complete surgical resection., Methods: The charts of 25 patients treated with neoadjuvant chemoradiation at Duke University Medical Center with biopsy-proven, locally advanced adenocarcinoma of the pancreas were reviewed. Tumors were defined as locally advanced based on radiographic or intraoperative evidence of disease that abuts the superior mesenteric artery or vein (n = 22) or involves lymph nodes that are within the proposed radiation field (n = 3). All 25 patients received external beam radiotherapy (median dose 4500 cGy) in daily fractions of 180 cGy over 5 weeks. All patients concurrently received 5-fluorouracil (FU), and many also received mitomycin C or cisplatin, or both. Patients were given a 3- to 4-week break before a restaging computed tomographic (CT) scan was performed. Three patients were not restaged: one died from metastatic disease; one was reclassified as having a neuroendocrine tumor; and one was lost to follow-up., Results: On restaging after neoadjuvant therapy, 64% of patients had stable or decreased primary tumor size. Radiographically, two patients appeared potentially resectable, and seven others developed evidence of metastatic disease. Eight patients underwent exploration, but only five could be resected. Of the five patients resected, only one had negative margins and negative lymph nodes. This patient had significant pancreatitis on initial exploration. After neoadjuvant therapy, he had a complete response radiographically, and there was no residual cancer in his resection specimen. Pathologic examination of the other resection specimens suggested that despite significant tumor fibrosis, malignant cells persist even at the periphery of the lesions., Conclusion: Although neoadjuvant chemoradiation has many theoretical advantages in managing pancreatic malignancy, true pathologic downstaging of locally advanced lesions into tumors that can be removed with negative nodes and margins appears to be a rare event with currently used therapeutic regimens.

Objective: The purpose of this study was to investigate the role of positron emission tomography (PET) with 18F-fluorodeoxyglucose (FDG) in differentiating benign from malignant disease in patients with possible pancreatic malignancy., Subjects and Methods: All patients with a possible diagnosis of pancreatic carcinoma based on CT or ERCP findings were eligible for inclusion in this prospective study. PET imaging of the abdomen was performed in 37 patients and was interpreted as positive if FDG activity in the pancreas exceeded background activity and as negative if activity was less than or equal to background activity. Semiquantitative analysis was performed by calculating a standardized uptake ratio. Studies were reviewed independently by two radiologists, and results were correlated with biopsy results and with CT and ERCP findings. Sensitivity and specificity of FDG PET for revealing pancreatic malignancy was determined., Results: FDG activity in the pancreas was increased in 24 patients, and adenocarcinoma was diagnosed in 22 of these patients (92%). Two patients (8%) with increased activity had benign disease, including one patient with chronic pancreatitis who showed no evidence of tumor at laparotomy and one patient with a mucinous cystic tumor who showed no malignant features at laparotomy. FDG uptake was low or normal in 13 patients, 10 of whom (77%) had benign disease. FDG uptake was also low in three patients with adenocarcinoma, whose tumor size ranged from 2 to 4 cm in diameter. The mean standardized uptake ratio value for malignant disease was 5.1 (range, 1.0-10.1) and for benign disease was 1.9 (range, 0.0-5.8) (p < .001). The sensitivity of FDG PET for revealing malignant disease in the pancreas was 88% and the specificity was 83%., Conclusion: FDG PET is a sensitive and specific noninvasive technique for the diagnosis of pancreatic malignancy.

Purpose: The purpose of this study was to analyze interobserver agreement in the interpretation of unenhanced helical CT (UHCT) for the evaluation of ureteral stone disease and obstruction., Method: One hundred three UHCT examinations were independently and retrospectively reviewed by five readers including attending radiologists, a radiology resident, and an attending urologist. Examinations were interpreted as positive, negative, or indeterminate for ureteral stone disease and obstruction. The Cohen kappa test was used to measure interobserver agreement. The accuracy of the readers was also assessed., Results: The kappa value ranged from 0.67 to 0.71 among the three attending radiologists and from 0.65 to 0.67 among the radiology attending physicians and radiology resident. Although the urologist tended to agree less well with the other readers (kappa range: 0.33-0.46), there was no statistically significant difference (p < 0.05) in the accuracy among all five readers. The percentage of cases interpreted as indeterminate ranged from 8 to 25% and almost invariably involved difficulty distinguishing phleboliths from minimally obstructing distal ureteral calculi. The percentage of UHCT scans correctly interpreted as positive and correctly interpreted as negative ranged from 73% (n = 27) to 86% (n = 32) and 63% (n = 22) to 86% (n = 30), respectively., Conclusion: Interobserver agreement was very good among the radiology attending physicians and resident and moderate with the urologist. The examination is an accurate technique in the evaluation of ureteral stone disease, although limitations exist, particularly in the diagnosis of minimally obstructing distal ureteral calculi.

Objective: The purpose of this study was to determine the optimal time for obtaining delayed images with contrast-enhanced CT in patients who have intrahepatic or hilar cholangiocarcinoma., Subjects and Methods: CT studies were performed in 25 consecutive patients with proven cholangiocarcinoma, including six patients who had undergone radiotherapy or chemotherapy. Dynamic images of the liver were obtained after 150 ml of IV contrast material was administered at 3 ml/sec. Delayed CT images were then obtained at 10, 20, and 30 min. Tumor-liver attenuation difference was determined quantitatively for each time period. Images were qualitatively evaluated by three observers for attenuation of the tumor (hypoattenuating, isoattenuating, or hyperattenuating) relative to the liver. Observer confidence for tumor detection was graded on a four-point scale. Dynamic and delayed images were compared for tumor conspicuity., Results: On dynamic images, 18 tumors (72%) were hypoattenuating, six (24%) were isoattenuating, and one was heterogeneous. On delayed images, 15 (60%) of these 25 tumors were isoattenuating and nine (36%) were hyperattenuating compared with the liver. Tumor-liver attenuation difference was greatest on dynamic studies (p < .01) and did not differ significantly among the three delay times (p > .20). All tumors seen on delayed images were also seen on dynamic images; however, in three patients (12%), the confidence level for presence of tumor was better on delayed than on dynamic images. Confidence levels for presence of tumor did not vary significantly among the three delay times. Attenuation values on dynamic and delayed images did not differ for the groups of patients who had or had not undergone prior radiotherapy or chemotherapy (p > .05)., Conclusion: In the evaluation of hilar or intrahepatic cholangiocarcinoma, delayed CT images are helpful for tumor characterization and may improve observer confidence for the presence of tumor. The optimal time for acquisition of delayed images is 10-20 min after contrast media injection.

Background: Studies of the influence of human leukocyte antigen (HLA) matching on cardiac transplant outcome have proved inconclusive, mainly due to the lack of well-matched grafts. However, a growing number of studies report improved clinical course and patient survival in cases with increased HLA compatibility. Opelz et al. believe these benefits justify the introduction of prospective HLA-matching strategies., Methods: We performed univariate and multivariate analyses to examine the short- and medium-term influence of HLA matching on 556 consecutive primary heart transplants performed at a single center between 1983 and 1994. Overall graft survival at 1, 3, and 5 years was 80%, 74%, and 67% respectively. Sixteen (2.9%) grafts failed within 5 days and were not considered in the analysis of the HLA matching and graft survival data., Results: Complete HLA-A, -B, and -DR typing data were available on 477 transplant pairs. The results demonstrate a 12% 1-year survival advantage for 31 patients with zero to two HLA antigen mismatches compared with three to six mismatches. The influence of each individual locus was 6.1%, 8.4%, and 5.4% for zero HLA-A, -B, and -DR mismatches, respectively, compared with two mismatches. However, when outcome from 1 to 5 years was considered, analysis of the role of each locus revealed marked differences. HLAA-matched grafts (n=45) had a 24% lower survival rate compared with two-antigen-mismatched grafts (n=148; 88% [SE 3.1] vs. 64% [SE 8.2], respectively; P=0.009). Furthermore, 34% of HLA-A-matched grafts failed between 1 and 5 years, compared with only 5% of HLA-B-matched grafts (P=0.013)., Conclusions: These data suggest that although HLA matching is effective at reducing acute graft loss, in the longer term, HLA-A matching may impair survival. HLA-A may serve as a restriction element for indirect presentation of allopeptides or tissue-specific minor histocompatibility antigens, facilitating chronic graft loss. Therefore, we advocate a differential role for HLA matching over two epochs. A blanket approach to prospective matching for heart transplants may be premature for optimal long-term survival.

Purpose: Our purpose was to determine the diagnostic accuracy efficacy of a simple MR technique in differentiating hepatic hemangiomas from colorectal or breast metastases using a multiple reader method., Method: Thirty-seven cases with confirmed hepatic hemangiomas and 115 with confirmed hepatic metastases (colon primary, n = 86; breast primary, n = 29) evaluated with MRI at 1.5 T were retrospectively collected. A single lesion in a single slice from each patient was randomly selected; the images were masked and then were interpreted in random order by five separate readers blinded to the diagnosis using a five point diagnostic scale (from definite hemangioma to definite metastasis). Morphologic characteristics of lesion margin, signal intensity relative to other structures, and internal architecture (homogeneous versus heterogeneous) were also assessed independently of the five point diagnostic scale. Three of the readers had > 8 years of experience, while the other two had 1 and 3 years. The diagnostic scale results were subjected to receiver operating characteristic (ROC) analysis using a jackknife method. kappa-Statistics were applied to assess interreader agreement in the morphologic characteristics. A logistic regression model was used to determine which characteristics predicted pathology and reader diagnosis., Results: ROC analysis showed the average area under the curve over all readers was (0.91 (0.89-0.93 95% confidence interval) (p < 0.0001). An analysis of variance showed no significant difference between the areas under the curves of each reader (p = 0.6433). When the definite and probable categories for hemangioma and metastasis were combined, the sensitivity/specificity for the diagnosis of hemangioma ranged from 57 to 73%/91 to 97%. The positive/negative predictive value ranged from 72 to 84%/87 to 91%. For the morphologic assessment, there was significant agreement between the readers (p < 0.0001-0.0037). A sharp margin and lesion signal equal to or greater than CSF predicted the presence of a hemangioma (p = 0.0148 and p < 0.0001, respectively). A sharp margin, lesion signal equal to or greater than CSF, and a homogeneous internal architecture all predicted the reader diagnosis of definitely or probably hemangioma., Conclusion: For multiple readers, T2-weighted SE MRI alone is a very specific method for distinguishing hemangiomas from metastatic colon or breast carcinoma. Morphologic characteristics of a sharp margin and a high signal predict the presence of a hemangioma. Last, reader experience does not appear to have a significant effect on the specificity.

Malignant melanoma, a common malignancy whose prevalence is increasing, represents 1-3% of cancers in the United States [1]. At autopsy, metastatic deposits to the gut are frequently found, but less than 9% of melanoma patients are diagnosed with gastrointestinal metastases while living [2]. Modern management includes aggressive surgical therapy to prolong survival and to palliate the disease [3]. Therefore, imaging of metastatic melanoma is clinically important to detect extent and determine whether the patient would benefit by surgery. Gastrointestinal metastases may manifest as mucosal or submucosal masses, serosal implants, or carcinomatosis [4]. They arise more commonly in the mesentery or distal small bowel than the proximal gastrointestinal tract or colon. The purpose of this essay is to illustrate the appearance of melanoma metastatic to the gastrointestinal tract on luminal contrast studies and on CT and to emphasize the importance of early investigation of gastrointestinal symptoms in a patient with a history of malignant melanoma.

Rationale and Objectives: To control costs, it is increasingly important to make efficient use of imaging technology. We sought to determine and analyze the time required to complete each step of a body computed tomography (CT) scan, focusing on factors that influence patient throughput., Methods: Over 4 weeks, we prospectively monitored the time required for each step of a body CT scan (i.e., image time, check time, and clear time). Covariate data were collected by patient status: outpatient, inpatient, emergency department (ED), and intensive care unit (ICU); work shift; and radiologist training level (junior resident, senior resident, fellow, and attending). Technologists also predicted whether repeat images would be requested by the radiologist., Results: Three hundred eighty CT examinations were studied: 277 for outpatients, 90 for inpatients, 9 for ED patients, and 4 for ICU patients. The mean total examination time was 44.7 min (mean image time = 33.1 min, mean review time = 8.2 min, and mean clear time = 3.4 min), which did not differ significantly with patient status. A second opinion was sought from a consultant radiologist on the scans of 44 patients. Consultation was requested significantly more frequently (1) by junior residents than by senior residents or fellows and (2) for ED and ICU patients (22% and 50%, respectively) than in outpatients and inpatients (10% and 14%, respectively). Repeat images were obtained from 75 patients, and this was not significantly related to patient status, scan type, or radiologist training level. When the technologist predicted that no repeat images were needed, this prediction agreed with the radiologist in 86% of the cases. When the technologist predicted that repeat images were necessary, this prediction agreed with the radiologist in 56% of the cases., Conclusion: Reviewing scans before the patient leaves the CT suite adds considerably to the total time required to complete a scan, particularly if junior residents review scans. If technologists obtain repeat images at their discretion, time would be saved.

Significant technical advances have occurred in both computed tomography (CT) and magnetic resonance (MR) imaging that have resulted in improved visualization of the pancreas by both techniques. In this article we will consider the advances in both modalities and consider the relative roles of each in imaging the spectrum of pancreatic pathology.

Computed tomography is currently the standard diagnostic tool for the evaluation of the skull base. The complex anatomy of this area is the primary reason why planar bone scintigraphy is often unsatisfactory; exact localization of abnormalities may be very difficult. These limitations may be overcome by SPECT. Seventeen patients with clinical features of basal skull involvement were assessed by CT, SPECT, and planar scintigraphy. Subsequent clinical diagnoses were malignancy in 15 patients, vasculitis in 1 patient, and osteomyelitis in 1 patient. Computed tomography with IV contrast was performed through the skull base at 5 mm intervals. Planar scintigraphy with Tc-99m MDP was followed by SPECT. Bony involvement compatible with the clinical findings was demonstrated by CT scans in 6 patients, by planar scintigraphy in 7 patients, and by SPECT in 9 patients. The abnormalities that were identified by CT were all identified by SPECT. This study suggests that, in imaging the skull base, SPECT is more sensitive and provides better anatomical localization than planar imaging and appears useful in patients with a negative CT study.

The efficacy of topical formulations of acidic fibroblast growth factor (aFGF) in healing of full-thickness wounds has been studied in a diabetic db+/db+ mouse model. The effect of several formulation variables, dose, and application frequency was examined. It was found that wound healing in diabetic animals treated with aFGF or placebo was slower than in their nondiabetic littermates. The availability of aFGF from the viscous vehicle employed in this study (1% hydroxyethyl cellulose) was demonstrated in vitro using diffusion cells. The viscous formulation of aFGF was equally effective in wound healing as a nonviscous formulation in phosphate-buffered saline. A formulation containing heparin (necessary for full biological and conformational stability of aFGF) at a mass ratio of 3:1 to aFGF was more efficacious than formulations with lower heparin: aFGF ratios. Wounds treated with three doses of 3.0 micrograms/cm2 aFGF healed faster than those treated with a single dose of 3.0 micrograms/cm2 aFGF. Three applications of 3.0 or 0.6 microgram/cm2 a FGF were equally effective in accelerating wound healing.

The design of an aqueous formulation for acidic fibroblast growth factor (aFGF) requires an understanding of the type of compounds that can either directly or indirectly stabilize the protein. To this end, spectrophotometric turbidity measurements were initially employed to screen the ability of polyanionic ligands, less specific compounds, and variations in solution conditions (temperature and pH) to stabilize aFGF against heat-induced aggregation. It was found that in addition to the well-known protection of aFGF by heparin, a surprisingly wide variety of polyanions (including small sulfated and phosphorylated compounds) also stabilizes aFGF. These polyanionic ligands are capable of raising the temperature at which the protein unfolds by 15-30 degrees C. Many commonly used excipients were also observed to stabilize aFGF in both the presence and the absence of heparin. High concentrations of some of these less specific agents are also able to increase the temperature of aFGF thermal unfolding by as much as 6-12 degrees C as shown by circular dichroism and differential scanning calorimetry. Other compounds were found which protect the chemically labile cysteine residues of aFGF from oxidation. Aqueous formulations of aFGF were thus designed to contain both a polyanionic ligand that enhances structural integrity by binding to the protein and chelating agents (e.g., EDTA) to prevent metal ion-catalyzed oxidation of cysteine residues. While room-temperature storage (30 degrees C) leads to rapid inactivation of aFGF in physiological buffer alone, several of these aFGF formulations are stable in vitro for at least 3 months at 30 degrees C. Three aFGF topical formulations were examined in an impaired diabetic mouse model and were found to be equally capable of accelerating wound healing.

Monoclonal antibodies that target T cells have shown some benefit in rheumatoid arthritis although responses have not been long lasting. This is partly due to insufficient therapy consequent upon antibody immunogenicity. Use of humanised antibodies, which are expected to be less foreign to man than conventional rodent antibodies, might overcome this problem. We therefore assessed in a phase 1 open study the potential of a "lymphocyte depleting" regimen of the humanised monoclonal antibody CAMPATH-1H in 8 patients with refractory rheumatoid arthritis. Apart from symptoms associated with first infusions of antibody, adverse effects were negligible. Significant clinical benefit was seen in 7 patients, lasting for eight months in 1. After one course of therapy, there was no measurable antiglobulin response, although 3 out of 4 patients have become sensitised on retreatment. Humanisation reduces the immunogenicity of rodent antibodies but anti-idiotype responses may still be seen on repeated therapy, even in patients sharing immunoglobulin allotype with the humanised antibody.

ANCA are markers for systemic vasculitis such as Wegener's granulomatosis (WG) and microscopic polyarteritis (MPA) and are usually of the IgG isotype. However, IgM ANCA may rarely occur in isolation, and in these circumstances, we have found that they are associated clinically with a syndrome of pulmonary hemorrhage and systemic vasculitis. How frequently IgM ANCA may occur in conjunction with IgG has not previously been investigated. We report here a study of 24 consecutive patients with IgG ANCA-positive systemic vasculitis (14 WG, 10 MPA) in whom we determined whether IgM ANCA occurred in association with IgG ANCA, and if so, whether this had clinical importance. Eight patients were found to have IgM ANCA as well as IgG ANCA, and of these, seven presented with severe pulmonary hemorrhage. None of the IgM ANCA-negative patients presented with pulmonary hemorrhage. Although the occurrence of pulmonary hemorrhage in ANCA positive vasculitis was closely correlated with the presence of IgM ANCA, the antigen specificity of these IgM autoantibodies was variable, since both myeloperoxidase (4 patients), PR3 (3 patients), and an unknown ANCA antigen (1 patient) were found to be targets. We conclude that knowledge of ANCA isotype may have important clinical and therapeutic implications for the management of patients with systemic vasculitis.

A cystic fibrosis (CF) clinic for adults was established in 1977. We have reviewed the data on 164 patients who attended between 1977 and 1989. Twenty four patients had died, 11 being over 20 years of age at the time of death. Of the 140 patients still alive, 61% were male and 53% were aged over 20 years. Only 55% were diagnosed by one year and 88% by ten years. Almost all patients had respiratory symptoms and sputum culture yielded pseudomonas species in 69%. Other respiratory problems included major haemoptysis and pneumothorax, each in 10%. We found a wide range of respiratory impairment among older patients. Among 33 patients aged over 23 years, the mean (+/-S.D.) percent predicted FEV1 and FVC were 53.3% (+/- 18%) and 71.4 (+/- 20%) respectively. Mean weight in this group was 92.5% (+/- 14) of predicted. Malabsorption occurred in most patients and meconium ileus equivalent occurred in 34%. Other complications were clinical hepatomegaly (16%), diabetes mellitus (9%) and arthropathy (20%). Most patients were taking continuous antibiotics by mouth (89%) and by nebuliser (48%), beta-2 agonists by inhaler (57%) and oral steroids (29%). Almost all were taking multivitamins, pancreatic replacement therapy and multiple nutritional supplements. The number of CF "bed days" grew 12 fold since 1979 and the mean stay in hospital was double the hospital mean. The economic impact was such that over 1/4 of the annual hospital antibiotic budget was expended on CF patients.

The improved survival of cystic fibrosis (CF) patients is partly due to intensive treatment for their chronic infections. Treatment usually includes intravenous and nebulised aminoglycoside antibiotics and they receive a large cumulative dose of these antibiotics over their lifetime. There is little information in the literature on the prevalence of ototoxicity due to aminoglycoside in these patients. We performed pure tone audiometry on 43 CF patients aged 14-42 years. Seven (16%) had bilateral sensorineural hearing loss (SNHL) for high frequency sounds, consistent with aminoglycoside induced ototoxicity. However, only 2 of these patients had documented toxic serum levels in the past. The identification of bilateral SNHL in one in six adult CF patients is a cause for concern. It may be that the high cumulative dose of aminoglycosides received by these patients may be causing inner ear injury in the absence of specific episodes of toxic serum levels.

Anti-neutrophil cytoplasm antibodies (ANCA) are specific markers for systemic vasculitis. In view of the autoreactivity to other autoantigens reported in patients with monoclonal immunoglobulins (MIg), the reactivity of 150 sera from 125 patients with MIg was tested for ANCA by radioimmunoassay (RIA) with inhibition stage and indirect immunofluorescence (IIF). Seven were positive for IgG ANCA, all with IgG MIg and 5 were positive for IgM ANCA, 4 with IgM MIg and 1 with IgG MIg. No IgA ANCA were found. The patterns seen on IIF were identical to those seen with sera from patients with systemic vasculitis and were cytoplasmic in 6 and peri-nuclear in 6. The restriction of the ANCA activity to the MIg was studied in six sera by light chain specific RIA, and anion exchange fractionation of the sera. The ANCA activity appeared to be polyclonal in at least three sera and could be found in the monoclonal fraction in only three patients. Associated autoimmune diseases were found in some of these ANCA positive patients including Sjögren's syndrome, MacDuffie hypocomplementemic vasculitis and rheumatoid polyarthritis but the classical vasculitic features normally associated with ANCA were not observed. We conclude that ANCA is a further autoreactivity present in some sera with MIg and discuss the relation between monoclonal gammopathies and autoimmunity.

Atypical mycobacterial disease has been described in a small number of patients with cystic fibrosis. Apart from one uncontrolled study, there is little information regarding atypical mycobacterial skin reactivity in this group of patients. We evaluated delayed cutaneous hypersensitivity to purified extracts of Mycobacterium avium, Mycobacterium intracellular, Mycobacterium kansasii and Mycobacterium bovis in 23 healthy controls and 43 adult and adolescent patients with cystic fibrosis. Fifteen of the cystic fibrosis group were receiving regular corticosteroids. Additionally, direct smear examination and Lowenstein Jensen culture were performed on sputum from the cystic fibrosis group. The prevalence of positive skin reactions was similar in the group with cystic fibrosis (30%) and in the control group (57%). Subgroup analysis showed that those cystic fibrosis patients receiving corticosteroids had a markedly lower prevalence of positive reactions (7%) compared to controls (P less than 0.01). When this subgroup was excluded from analysis, the prevalence of positive skin reactions among patients with cystic fibrosis was 43%. In the prospective sputum bacteriology study, one of the 43 cases grew Mycobacterium avium-intracellulare and had clinical and radiological evidence of this disease. Of note, this patient showed positive skin tests to all four mycobacterial species tested. Our data show no difference in the prevalence rate of positive skin reactions to atypical mycobacterial antigens between a control population and an adult cystic fibrosis population. In addition, the predictive value of skin testing is low in cystic fibrosis due to the high prevalence of cross-reactivity between different mycobacterial species and the high prevalence of anergy among those patients with advanced disease receiving treatment with corticosteroids.