Analysis of surgical complications and patient outcomes in a residency training program.

OBJECTIVES: This study compares surgical complications and patient outcomes between pelvic reconstructive surgery performed by an experienced surgeon (group 1) and those performed by resident physicians with the senior surgeon assisting and teaching (group 2). STUDY DESIGN: During a 5-year interval, 310 consecutive women underwent vaginal prolapse repair. Demographic, historic, and preoperative physical examination variables were compared. Intraoperative and postoperative outcomes were also compared. RESULTS: Patients operated on by the senior surgeon (Bob L. Shull) were thinner (group 1 vs group 2: 25.8 kg/m2 vs 27.1 kg/m2; P =.014), more often had prior prolapse or incontinence procedures (55% vs 33%; P <.001), and required shorter operating times (124 minutes vs 140 minutes; P =.002). The senior surgeon's patients differed from the resident physicians' patients with regard to stage of pelvic organ prolapse. No differences were observed for patient age (P =.51), estimated blood loss (P =.50), urologic complications (P =.59), and hospital stay (P =.25). The durability of the repairs was not different between the groups. CONCLUSIONS: We have demonstrated that in a tertiary referral practice resident surgeons can be taught to perform complex vaginal surgery with the only observed disadvantage being a slightly prolonged operative time.

A transvaginal approach to repair of apical and other associated sites of pelvic organ prolapse with uterosacral ligaments.

OBJECTIVE: The objectives of this study were (1) to describe a group of women with pelvic organ prolapse associated with apical loss of support through grading with the Baden-Walker halfway system before, during, and after the corrective operation, (2) to describe the operative repair of the support defects, (3) to report the morbidity associated with the operative repair, and (4) to assess the durability of the repair at each site. STUDY DESIGN: Between January 1, 1994, and December 31, 1998, a total of 302 consecutive women with apical and associated other support defects were evaluated before, during, and after the corrective operation by the senior author (Bob L. Shull). All patients underwent transvaginal reconstructive surgery with native tissue. Two hundred eighty-nine patients (96%) returned for at least one postoperative visit, and they constitute the group used for the follow-up data. Perioperative morbidity was considered to include hemorrhage necessitating homologous blood transfusion, visceral injury, neurologic impairment, or death. Durability was assessed by means of life-table analysis for each of 5 sites in the vagina. RESULTS: All patients had preoperative or intraoperative evidence of grade 1 or greater apical loss of support of and at least one other site of pelvic organ prolapse. Two hundred eighty-nine patients (96%) returned for at least one postoperative visit. Two hundred fifty-one patients (group 1, 87%) had optimal anatomic outcomes, with no persistent or recurrent support defects at any site. Thirty-eight patients (group 2, 13%) had one or more sites with at least grade 1 loss of support during the follow-up interval. Twenty-four of these 38 patients had grade 1 defects that were detectable only on careful pelvic examination. Fourteen of these patients (5%) had grade 2 or greater persistent or recurrent support defects. The anterior segment (bladder) was the site with the most persistent or recurrent support defects, which means that it was the site of the least durable repair. The urethra and cuff had the most durable repairs. Morbidity included a 1% transfusion rate, a 1% ureteral injury or ureteral kinking rate, and a 0.3% postoperative death rate. CONCLUSION: Careful preoperative and intraoperative evaluation of pelvic support defects and the use of native connective tissue and uterosacral ligaments are associated with excellent anatomic outcomes. The durability of the surgical correction varies according to the individual site of repair and the duration of postoperative follow-up.

Accuracy of clinical assessment of paravaginal defects in women with anterior vaginal wall prolapse.

OBJECTIVE: The objective of this study was to determine the accuracy of clinical assessment of paravaginal defects in women with anterior vaginal wall prolapse. STUDY DESIGN: A retrospective chart review of all women undergoing surgery for anterior vaginal wall prolapse during the years of 1994 to 1996 identified operative notes that described the surgical assessment of paravaginal support. These surgical findings were compared with the preoperative clinical assessment. Clinical parameters that predicted poor correlation were identified. Statistical analysis used the chi(2) test. RESULTS: One hundred seventeen patients had surgery for anterior vaginal prolapse. Seventy had documentation of an intraoperative paravaginal support evaluation. Of these, 44 patients had vaginal procedures, and 26 had abdominal procedures. All patients had at least stage 2 prolapse before surgery, and all were noted to have excellent pelvic support 4 to 6 weeks after surgery. The prevalence of paravaginal defects at surgery was 47% on the right and 41% on the left. The sensitivity and negative predictive value for the clinical assessment for paravaginal defects were good on both the right and left sides, whereas the specificity and positive predictive values were poor. Stage of prolapse, previous hysterectomy, or previous anterior colporrhaphy did not significantly affect the accuracy of the clinical examination in predicting fascial defects. However, previous retropubic urethropexy did significantly decrease the accuracy of the clinical examination in predicting right paravaginal defects (P <.01) but not left. CONCLUSION: Although preoperative clinical assessment for paravaginal defects is useful, it does not substitute for careful intraoperative evaluation for endopelvic fascial defects.

Urinary incontinence and pelvic organ prolapse in nulliparous women.

OBJECTIVE: To determine differences between vaginally parous and nulliparous women presenting with urinary incontinence and pelvic organ prolapse. METHODS: Seven hundred forty eight consecutive referrals with urinary incontinence or pelvic organ prolapse, 62 of whom were nulliparous, were included in the analysis. Five hundred thirty-seven (72%) had urinary incontinence and 235 (31%) had at least stage III pelvic organ prolapse. Each subject had standard history, physical examination, and multichannel urodynamic testing. Differences between parous and nulliparous women were compared using parametric and nonparametric analysis of variance and the chi2 test with Yates correction where appropriate. RESULTS: The only significant demographic difference between the groups was that parous women had more previous continence and prolapse surgery. There were significant differences in distribution of diagnoses according to parity, with the nulliparas much less likely to have pelvic organ prolapse. Among incontinent women without prolapse, nulliparas were significantly more likely to have pure detrusor instability. Of those with pure genuine stress incontinence, nulliparas were older, had less anterior vaginal wall descent, less bladder neck mobility, narrower genital hiatus and perineal body measurements, and lower maximum urethral closure pressures. Of those with pure detrusor instability, the only difference was that nulliparas were significantly younger. For women with stage III pelvic organ prolapse or worse, no significant difference in any measured characteristic was noted. CONCLUSION: Nulliparous women were less likely to present with pelvic organ prolapse and those with urinary incontinence differed little from incontinent parous women.

Urethral prolapse after collagen injection.

A woman who was treated for intrinsic urethral sphincteric deficiency with periurethral injection of glutaraldehyde cross-linked collagen had prolapse of the urethral mucosa and recurrence of incontinence. She subsequently required surgical resection and a fascia lata sling. This is the first known occurrence of this postinjection complication.

Physiologic evaluation of the pelvic floor.

This article has addressed the value, technique, rationale, and limitations of the commonly performed physiologic tests of the pelvic floor. Urodynamics provides a means for evaluation of the lower urinary tract and for assessment of the filling and emptying phases of the bladder. Neurophysiologic tests including EMG and nerve conduction studies offer methods to assess the neuromuscular integrity of the urethral and anal sphincteric mechanisms and the pelvic floor. Anorectal manometry studies provide a means of measuring pressure in the rectum and anal canal, rectal compliance, and anorectal reflexes and sensation in patients with anal incontinence and select patients with constipation. Colonic transit studies permit an assessment of functional constipation and may be helpful in the evaluation of patients in whom standard management of constipation has failed. Any test used in the evaluation of patients with pelvic floor dysfunction should be validated and found to be reliable. Most patients in need of complex physiologic evaluation are identified on the basis of their history and physical examination findings.

Clinical predictors of urinary incontinence in women.

OBJECTIVE: Our purpose was to characterize historic and clinical parameters in incontinent women to determine the predictive value for urodynamic diagnoses. STUDY DESIGN: The analysis includes 535 consecutive women with final diagnoses of genuine stress incontinence, detrusor instability, or both. Evaluations included a standardized history, examination, urinary diary, quantitation test, and urodynamics. The analysis used one-way analysis of variance, chi 2 analysis with Yates' correction, and Fisher's exact test. RESULTS: A total of 351 (66%) women were diagnosed with genuine stress incontinence, 102 (19%) with detrusor instability, and 82 (15%) with both. Half had symptoms of both stress incontinence and urge incontinence, of whom only 21% had both genuine stress incontinence and detrusor instability. Fewer than half of women diagnosed with genuine stress incontinence or detrusor instability had just symptoms of stress incontinence or urge incontinence, respectively. Evaluation of historic, examination, and urinary diary data for their influences on the predictive value of pure stress incontinence or urge incontinence revealed statistical differences for urethral hypermobility, estrogen deficiency, and incontinent episodes, yet they were not clinically practical predictors. CONCLUSIONS: Pure symptoms identify fewer than half of patients with pure genuine stress incontinence or detrusor instability; historic and clinical parameters do not improve the sensitivity of these symptoms.

Diagnosing intrinsic sphincteric deficiency: comparing urethral closure pressure, urethral axis, and Valsalva leak point pressures.

OBJECTIVES: Our purpose was to compare three measures proposed to diagnose intrinsic sphincteric deficiency: maximum urethral closure pressure, Valsalva leak point pressure, and straining urethral axis. STUDY DESIGN: A total of 159 women with pure genuine stress incontinence had the three measures determined in a standardized fashion. Critical cutoff values for the Valsalva leak point pressure (52 cm) and urethral axis (22 degrees) were established by examining relative frequency distribution curves, using closure pressure of 20 as the arbitrary benchmark value for the prevalence of intrinsic sphincteric deficiency. The distribution of cutoff values is described and differences among the measures with respect to risk factors for intrinsic sphincteric deficiency and incontinence severity were determined. RESULTS: Half the subjects fell below at least one cutoff value, but only 10% fell below all three. Sixty-four percent of subjects with either low closure pressure or leak point pressure had low values for the other, whereas 21% had discordance between them. Only 53% of subjects with low closure pressure and 40% with low leak point pressure had an axis < or = 22 degrees. Conversely, a substantial portion (36%) of subjects with pure genuine stress incontinence without urethral hypermobility had neither low urethral or leak point pressures. All three cutoff-values were associated with risk factors for intrinsic sphincteric deficiency, but only low closure and leak point pressures had significant associations with the severity of incontinence. CONCLUSIONS: Intrinsic sphincteric deficiency should be diagnosed by a composite of historic, urodynamic, anatomic, and clinical severity criteria. We would include a maximum urethral closure pressure < or = 20, a Valsalva leak point pressure < or = 50, and a stress urethral axis < or = 20 in this composite.

Uroflowmetry in women with urinary incontinence and pelvic organ prolapse.

OBJECTIVE: To characterize uroflowmetry parameters in women with pelvic organ prolapse (POP) and urinary incontinence (UI) and to assess the effects of clinical and urodynamic variables on these parameters. PATIENTS AND METHODS: The study comprised 655 consecutive women who presented with UI or POP and who had interpretable uroflowmetry values. Normal uroflowmetry values were defined as a maximum flow (Q(max)) > or = 15 mL/s, a mean flow (Q(mean)) > or = 10 mL/s, a post-void residual volume (PVR) < or = 100 mL and a continuous, single-peak waveform. Parametric and non-parametric analysis of variance and chi-square analysis were used to compare differences between diagnostic groups. Multiple linear regression models were developed to evaluate factors considered to influence uroflowmetry. RESULTS: Of the 655 patients, 471 (72%) had UI of whom 16% had pure detrusor instability (DI), 69% pure genuine stress incontinence (GSI) and 15% with both, and 184 (28%) had POP, 26% of whom also had DI. Of all patients, 72% had normal uroflowmetry patterns, 13% had multiple peaks and 15% had patterns with interrupted flow; 56% had completely normal uroflowmetry. There were significant differences in uroflowmetry values between the POP and UI groups, with the former having a lower Q(max) and Q(mean) (P < 0.001), larger PVRs (P < 0.001) and a lower percentage of totally normal uroflowmetry (33% and 64%, respectively, P < 0.001). Of patients with POP, 30% had a PVR > 100 mL. Because of the differences, the POP and UI groups were evaluated separately in the regression analysis. In both groups, the most important determinants of flow rate were the volume voided and pressure transmission ratio (PTR). However, when several factors (including age, voided volume, PTR and maximum detrusor pressure with flow and at Q(max)) were included in the model, they accounted for only 23-26% of the variability of flow in the patients with UI and 36-39% of the variability in patients with POP. The subsets of patients with pure DI in both the UI and POP groups had higher PVR volumes than the other subsets. CONCLUSIONS: These results show that the positive correlation between flow rate and voided volume described in normal populations is also observed in women with UI and POP. However, most of the variability in urine flow was not attributable to factors such as age, voided volume and PTR, confirming the complexity of the micturition mechanism. Women with POP had more objective evidence of emptying-phase dysfunction than women with UI, although most emptied their bladders efficiently. Finally, the results suggest that women with DI exhibit dysfunction of both inhibitory and facilitory detrusor control.

The squirrel monkey as an animal model of pelvic relaxation: an evaluation of a large breeding colony.

OBJECTIVE: Findings of pelvic relaxation have been reported in up to 50% of older adult female squirrel monkeys. To evaluate further the potential use of the squirrel monkey as an animal model of pelvic relaxation, we objectively observed and described the perineal findings of 160 adult females. The aim of this study was to examine the relationship of perineal findings to age and parity, factors thought to predispose women to pelvic relaxation. STUDY DESIGN: The urethra, cervix, and anterior and posterior segments of the vagina were evaluated. The degree of support loss at each site was documented. Genital measurements were obtained by previously reported methods. The findings were tested for association with elements of obstetric history, age, and subspecies. RESULTS: The females represented three subspecies and ranged from 3 to 17 years old with parities of 0 to 10. The proportion of females with normal support was inversely related to increasing parity and age. Although birth weights, frequency of dystocia at term, and requirement for cesarean section did not differ significantly between females with and without evidence of prolapse, animals with multiple sites of prolapse tended to have infants with higher birth weights. Animals without prolapse were significantly younger and less likely to have been delivered of a term infant (p < 0.001). Subspecies differences unrelated to age or parity were found for each of the genital measurements. Differences were also found between animals with normal perineal findings and those with findings of prolapse. Animals with prolapse had shorter perineal bodies (p < 0.001), greater genital hiatal ratios (p < 0.001), and wider genital hiatal measurements (p < 0.001). Females with abnormal pelvic findings were of increased parity (4.0 vs 1.6, p < 0.001) and age (9.4 vs 6.3 years, p < 0.001) compared with those normal pelvic findings. CONCLUSION: Analysis of genital prolapse in a large population of breeding squirrel monkeys demonstrated an association of loss of pelvic support with age and parity. A tendency for loss of support at multiple sites was associated with obstetric complications. These observations support continuing investigation into the nature and cause of spontaneous pelvic relaxation in this species and support the potential use of this nonhuman primate as an animal model.

The squirrel monkey: an animal model of pelvic relaxation.

OBJECTIVE: Casual observations of pelvic prolapse have previously been made in adult female squirrel monkeys. We objectively observed and described the pelvic findings of the adult females in our colony to evaluate their potential as an animal model for studies of pelvic relaxation. STUDY DESIGN: Subsets of a group of 28 adult female squirrel monkeys were examined on three separate occasions and photographed to document perineal findings. Four areas of loss of support in the vagina were described, including the urethra, anterior segment, cervix, and posterior segment. The degree of loss of support at each specific site was documented. Also, three measurements were obtained from each animal, including the distance from the base of the clitoris to the posterior edge of the perineal body and the length and width of the introitus. The length of the perineal body was calculated. RESULTS: Fifty percent of the animals had at least one site of pelvic prolapse at rest or on straining. With straining 50% had urethral prolapse, 21% had anterior segment prolapse, and 11% had posterior segment prolapse. None of the animals had evidence of cervical prolapse. Of the animals with findings of pelvic relaxation, 43% had combined prolapse at two or more sites. Of the measurements obtained, only the width of the introitus was significantly different between the group of normal animals and those found to have pelvic prolapse (p < 0.0005). CONCLUSION: On the basis of observations of our colony of squirrel monkeys, a significant number of animals demonstrated findings of pelvic support loss with defects similar to those observed in humans. Although limitations to some techniques of evaluation and surgical intervention exist, the squirrel monkey is an animal model that may provide new opportunities for basic investigation of the pathophysiologic features and the management of pelvic relaxation.