Invazivni rak vrata maternice po učestalosti zauzima četvrto mjesto svih zloćudnih bolesti ženskog reprodukcijskog sustava, a godišnje se u Hrvatskoj otkrije 14 novooboljelih na 100.000 žena. Materijali i metode. Prospektivno je praćeno 120 bolesnica, liječenih u Klinici za ženske bolesti i porode KBC-a Zagreb i Medicinskog fakulteta Sveučilišta u Zagrebu tijekom razdoblja 1996–2002. godine zbog raka vrata maternice kliničkog stadija IB-IIB.Cilj. Odrediti podtip HPV-a metodom PCR-om i Inno Lippa-om kod bolesnica liječenih zbog invazivnog karcinoma vrata maternice, s posebnim osvrtom na miješane infekcije te utvrditi povezanost podtipa HPV-a sa histološkim tipom karcinoma vrata maternice. Rezultati. Za analizu je preostalo 102 bolesnice, prosječne dobi 51,38 ±14,04 (26–83) godine. U vremenu praćenja (5–98 mjeseci) umrlo je 23 (22,5%) bolesnice zbog raka vrata maternice. Statistički je značajna razlika u preživljenju bolesnica u odnosu na: infekciju HPV tipom 16 (p=0,0008) i postojanju miješane infekcije HPV-om (p=0,0134). Coxovom regresijom analizirali smo sve čimbenike, statistički značajne univarijantnom analizom, i formirali model za preživljenje bolesnica s rakom vrata maternice visoke predikcije (2=74,222, df=7, p, Invasive cervical carcinoma is the fourth of all neoplasms of female genital reproductive system and 14/ 100 000 women would got ill in Croatia during a year. Material and method. 120 patients with cervical carcinoma clinical stage IB-IIB treated in Department of Obstetrics and Gynecology Medical School University of Zagreb, were examined prospectivly during period 1996–2002. The aim of the study was to detect the HPV-type using PCR and Inno Lippa method and compare their significance as prognostic factors. Results. 102 patients left for the analysis, an average age was 51.38±14.04 (26–83). During the folow up of 5–98 months 23 (22.5%) patients died from cervical carcinoma. HPV type 16 infection (p=0.0008) and mixed HPV infection (p=0.0134) are connected with poor surviving. Using Cox regression model we analysed all the significant factors by univariate method and formed a model of high prediction for poor surviving (2=74.222, df=7, p