Sensitivity of the Papanicolaou smear in human immunodeficiency virus-infected women.

OBJECTIVE: To determine the sensitivity of the Papanicolaou smear in human immunodeficiency virus (HIV)-infected women. METHODS: Fifty-two HIV-seropositive women underwent colposcopy and had cytologic and histologic specimens obtained. Eighty-five consecutive women seen in our colposcopy clinic who stated that they were HIV-seronegative formed the comparison group. Demographic features of the groups and cytology and histology findings were compared. RESULTS: In HIV-infected patients, the prevalence of cervical dysplasia was 50%. An additional 15% had vulvar lesions alone. The sensitivity of cytologic smears for cervical dysplasia was 63% and specificity was 84%. These findings were not significantly different from those of a control group of women undergoing colposcopy for standard indications (P > .05). Twelve of 38 seropositive women for whom future appointments were scheduled were lost to follow-up. CONCLUSIONS: The sensitivity of the Papanicolaou test does not appear to be diminished in HIV-seropositive women. Screening colposcopy may be justified in view of the high prevalence of cervical and vulvar dysplasia, as well as the high noncompliance rate observed in this patient population.

Factors affecting perioperative pulmonary function in acute respiratory failure.

To determine the magnitude, duration, and associated factors of perioperative changes in pulmonary function, we retrospectively reviewed the medical records of 145 patients who required preoperative mechanical ventilation for acute respiratory failure before undergoing 200 surgical procedures. Patients were grouped into five pulmonary diagnostic categories: (1) adult respiratory distress syndrome (ARDS) (n = 49); (2) pneumonia (n = 20); (3) atelectasis (n = 65); (4) congestive heart failure (n = 11); and (5) acute ventilatory failure (n = 55). Sixty patients underwent intra-abdominal surgery, 135 patients required surgery on the periphery, and five patients had a thoracotomy. For all patients, PaO2/FIO2 declined significantly from 321 mm Hg (mean) preoperatively to 258 mm Hg intraoperatively, and shunt fraction (Qs/QT) increased from 0.16 to 0.23 without a significant change in PaCO2. The magnitude of the increase in Qs/QT did not differ among pulmonary diagnostic groups. Preoperatively, patients undergoing laparotomy had lower PaO2/FIO2 (278 vs 340) and higher Qs/QT (0.19 vs 0.14) than patients requiring surgery on the periphery. Intraoperatively, Qs/QT increased more during abdominal procedures than during peripheral procedures. Intraoperative hypoxemia (PaO2/FIO2 less than 80 mm Hg) occurred during 13 procedures. Hypoxemic patients had a mean increase in Qs/QT of 0.20 (0.25 preoperatively to 0.45 intraoperatively), and a significant increase in PaCO2 from 38 mm Hg to 45 mm Hg intraoperatively). In general, these patients had ARDS (n = 10), sepsis (n = 10), a laparotomy (n = 9), and intraoperative mechanical ventilation via the Ohio Anesthesia ventilator (n = 8), a commonly used operating room ventilator. Their preoperative peak airway pressure (54 cm H2O) and minute ventilation (20 L/min) requirements exceeded the capabilities of the Ohio Anesthesia ventilator and likely contributed to impaired gas exchange intraoperatively. Within the first several hours postoperatively, PaO2/FIO2 recovered to preoperative levels in all patients, even in those who had severe intraoperative hypoxemia develop and who underwent laparotomy. We conclude that most patients with acute respiratory failure receiving preoperative mechanical ventilation experienced mild-to-moderate deterioration in intraoperative pulmonary oxygen exchange that rapidly returned to preoperative levels after surgery. We recommend that necessary surgery not be postponed by concern that pulmonary function will be worsened by surgery and anesthesia.