Myomectomy fever: testing the dogma

OBJECTIVE

To test the hypothesis that elevated temperature is more common after abdominal myomectomy than after hysterectomy.

DESIGN

Retrospective cohort study.

SETTING

Academic medical center.

PATIENT(S)

One hundred one women who underwent abdominal myomectomy and 160 women who underwent total abdominal hysterectomy for benign disease from 1988-1993.

INTERVENTION(S)

Abdominal myomectomy.

MAIN OUTCOME MEASURE(S)

Temperature of > or = 38.5 degrees C within 48 hours after operation.

RESULT(S)

Although univariate analysis showed that the incidence of elevated temperature was slightly greater among patients who underwent myomectomy (33% versus 26%, relative risk 1.29, 95% confidence interval 0.88-1.90), multivariate logistic regression analysis showed a 3.29 relative risk of elevated temperature (95% confidence interval 1.56-6.96) with myomectomy after controlling for age, parity, estimated blood loss, and treatment by the general gynecology service.

CONCLUSION(S)

After controlling for confounders, myomectomy was found to be an independent predictor for fever in the first 48 hours after operation.

Comparison of ureteral and cervical descents during vaginal hysterectomy for uterine prolapse

OBJECTIVE

The study measured ureteral and cervical locations during vaginal hysterectomy for prolapse and the extent of parametrial ligament shortening possible.

STUDY DESIGN

Cervical and ureteral position were measured in 26 women undergoing uterine prolapse correction. Parametrial clamp tip location was also measured.

RESULTS

The cervix lay between 0 and -14.5 cm (below) the hymen (mean +/- SD -5.35 +/- 3.96 cm) and the ureters lay +5.0 to -4.0 cm (mean +/- SD +1.89 +/- 1.99 cm). Correlation of ureteral with cervical position was 0.69 (P <.01) and correlation with ipsilateral uterosacral ligament clamp positions was 0.80 (P <.01). Regression line slope relating cervical descent and cervix to ureter distance was 0.65, indicating that for every 3 cm of cervical descent there was 2 cm widening of the gap between the cervix and ureters and 1 cm descent of the ureter.

CONCLUSION

For every 3 cm of cervical descent the ureters descend 1 cm, thereby widening the ureterocervical gap and permitting ligament shortening during vaginal hysterectomy.

Posthysterectomy vaginal vault prolapse: primary repair in 693 patients

OBJECTIVE

To examine the results of primary repair of posthysterectomy vaginal vault prolapse in a current, large series of patients with long-term follow-up.

METHODS

From January 1976 to December 1987, 693 patients underwent primary repair of vault prolapse at the Mayo Clinic. The Mayo culdoplasty technique was used in 95% of these patients. Patients were followed up by reference to their Mayo Clinic medical records, a specifically designed questionnaire, and pelvic examination in a subgroup of patients.

RESULTS

The median age at operation was 66 years. Abdominal hysterectomy had been performed on 49.5% of patients and vaginal hysterectomy on 43.4% (hysterectomy type was not documented on 7.1%). The median number of years to vault prolapse repair after hysterectomy was 15.8 (range 0.4-48.4). Information about prolapse after primary repair was available for 504 patients (72.7%) and 80 had evidence or complaint (bulge, protrusion) of recurrent prolapse. Thirty-six of 693 patients (5.2%) had subsequent prolapse repair. Eighty-two percent of patients indicated satisfaction with the result. Complications of operation included entry into the bladder or rectum (2.3% of patients), vault hematoma (1.3%), cuff infection (0.6%), and ureteral complications (0.6%). The number of patients presenting for vault prolapse repair increased during the study interval.

CONCLUSION

The Mayo culdoplasty can be performed with minimal morbidity. It achieved an anatomic restoration of upper vaginal support in a high percentage of patients with long-term follow-up.

Avoiding serious infections associated with abdominal hysterectomy: a meta-analysis of antibiotic prophylaxis

OBJECTIVE

Our objective was to determine whether the use of preoperative antibiotics prevents serious infections associated with total abdominal hysterectomy.

STUDY DESIGN

We identified 25 randomized controlled trials of antibiotic prophylaxis that used rigorous protocols. We performed meta-analysis and cumulative meta-analyses for all of the trials, and then we performed separate meta-analysis for cefazolin, metronidazole, and tinidazole.

RESULTS

Overall, 21.1% (373 of 1768) of the patients who did not receive antibiotic prophylaxis had serious infections after abdominal hysterectomy. Among patients who received any antibiotics, we found that 9.0% (166/1836) had serious postoperative infections; among those who received cefazolin, metronidazole, or tinidazole, 11.4% (70 of 615), 6.3% (17 of 269), and 5.0% (5 of 101), respectively, had serious postoperative morbidity. The differences in the prevalence of infection between women who received prophylaxis and women who did not receive prophylaxis were statistically significant (any antibiotics, p = 0.00001; cefazolin, p = 0.00021; metronidazole, p = 0.015; and tinidazole, p = 0.034).

CONCLUSION

Because preoperative antibiotics are highly effective in the prevention of serious infections associated with total abdominal hysterectomy, we believe they should be used routinely. In addition, we believe that the use of controls who receive no treatment is no longer justified in trials of antibiotic prophylaxis for total abdominal hysterectomy.

A hypothesis to explain paradoxical racial differences in neonatal mortality

We suggest an explanation for a pediatric paradox: the low-birthweight-specific (LBW) neonatal mortality rate (NMR) for black infants is less than the NMR for white infants, even though the overall NMR for black infants is about twice the overall NMR for white infants. Analyzing the Delivery Interview Program data set--a large matrix of information collected at the Boston Hospital for Women (now the Brigham and Women's Hospital)--we found that, overall, black infants' gestations are about four days shorter than white infants'. However, after stratifying by birthweight, we found a reversal in our data, namely, LBW (< 2,500 g) black infants' gestations are seven days longer than the gestations of LBW white infants. We believe that this increased chronological maturity may account for some of the survival advantage of the LBW black infant compared with the LBW white infant of the same weight.

Anatomy of anal sphincters and related structures in continent women studied with magnetic resonance imaging

Five anally continent nulliparas of reproductive age were studied with magnetic resonance imaging. The internal and external anal sphincters could be easily delineated, as could the intervening longitudinal musculature, puborectalis muscle, anococcygeal raphe, anorectal lumen, vagina, uterus, bladder, urethra, coccyx, and pubis. The shape of the sphincters was nearly cylindrical, with an anterior component averaging 18.3 mm thick and 28.0 mm long. Fifty-four percent of this anterior thickness was attributable to the internal sphincter. The anorectal angle varied considerably, with a mean of 86.8 +/- 19.1 degrees (range 60-112). The angle between the portion of the rectal lumen supported by the anococcygeal raphe, or levator plate, and the plane of the puborectalis muscle was consistent at 149.0 +/- 6.3 degrees (138-154). The finding of anterior anal sphincters with substantial thickness and length contrasts markedly with a view often pictured in the literature of a female anal sphincter that narrows anteriorly to half its posterior length and forms a small bundle of muscle rather than a broad band. Knowledge of these relationships is important in primary repair of obstetric sphincter lacerations as well as in surgical correction of anal incontinence.

Survival rates of monoamniotic twins do not decrease after 30 weeks' gestation

A search of pathology records from the years 1967 to 1988 at Women and Infants' Hospital of Rhode Island (138,232 live births) revealed 24 sets of histologically confirmed monoamniotic twins. All records were available for review. Among the 17 sets of monoamniotic twins that reached 30 weeks' gestation with at least one twin still alive, there were no further fetal deaths. The risks of early delivery in these pregnancies appear to outweigh the risk of fetal death as a result of monoamniotic status alone. These data do not show an advantage to early delivery.