Cancer-associated human papillomavirus types are selectively increased in the cervix of women in the first trimester of pregnancy.
OBJECTIVE: Because of incomplete data in the current literature, we sought to determine whether pregnancy is an independent risk for the detection of human papillomavirus infection. STUDY DESIGN: Pregnant patients in their first trimester who are seen for routine care at an obstetrics clinic and nonpregnant patients receiving routine care at a sexually transmitted diseases clinic and a gynecology clinic were recruited. Cervical cells were collected by saline solution lavage, and deoxyribonucleic acid was extracted and tested for the presence of human papillomavirus deoxyribonucleic acid with the hybrid capture assay. This assay detects the deoxyribonucleic acid of five "low cancer risk" and nine "high cancer risk" human papillomavirus types with two separate pools of deoxyribonucleic acid probes. RESULTS: Results from 245 pregnant women, 248 patients from the sexually transmitted diseases clinics, and 246 gynecology clinic patients were analyzed. Human papillomavirus deoxyribonucleic acid was detected in samples from 31% of the pregnant patients compared with 17.7% and 18.6% of the sexually transmitted diseases clinic and gynecology clinic patients, respectively. There was no significant difference in positivity for the "low-risk" human papillomavirus types among the three groups (positivity rates from 8.9% to 12.7%), but the pregnant patients had a significantly higher positivity rate for the "high-risk" human papillomavirus types (24.9% compared with 13.3% and 11.4% for the sexually transmitted diseases and gynecology clinic patients, respectively; p < 0.001). A multiple logistic regression analysis showed that pregnancy was an independent predictor of a positive test result for a "high-risk" human papillomavirus type (odds ratio 1.79, 95% confidence interval 1.11 to 2.89) but not a positive test result for a "low-risk" type. CONCLUSIONS: Because sexual activity was not greater among the pregnant patients, we propose that the increased detection of "high-risk" human papillomavirus types among the pregnant patients represents a selective activation of these viruses by hormonal or immunologic factors associated with pregnancy.