Evaluation of Postpartum Pelvic Floor Physical Therapy on Obstetrical Anal Sphincter Injury: A Randomized Controlled Trial

Abstract

Objective:

To evaluate a short course of postpartum pelvic floor physical therapy (PFPT) on symptoms and quality of life after obstetrical anal sphincter injuries (OASIS).

Methods:

This was a multicenter, randomized controlled trial. Adult women were eligible if they had a vaginal delivery complicated by OASIS and did not have a prior vaginal delivery after 24 weeks’ gestation. Participants were randomized to standard care plus PFPT or standard care alone. The primary outcome was change in the Pelvic Floor Distress Inventory (PFDI-20) at 12 weeks postpartum. Secondary outcomes included other validated questionnaires assessing pelvic floor symptoms and bother.

Results:

We approached 89 eligible women; 50 were enrolled and randomized. We excluded 6 participants in the PFPT group and 1 in the standard care group who withdrew before study initiation or did not provide baseline data. Thus, we analyzed 19 in the PFPT group and 24 in the standard care group. The groups were similar with regard to most baseline characteristics.

The PFPT group reported significant improvement in pelvic floor symptoms and bother compared with the standard care group, reflected by a decrease of 29.2 (IQR: −58.3, −4.2) compared to 0.0 (IQR: −14.6, 20.8) on the PFDI-20 from 2 weeks to 12 weeks postpartum (p=0.002). Significant differences also were observed for all PFDI-20 subscales (all p≤0.02).

Conclusion:

At 12 weeks postpartum, the PFPT group reported a significant improvement in pelvic floor symptoms and bother compared to the standard care group.

Clinical Trial Registration:

Clinical.Trials.gov, NCT02304016

PRECIS:

At 12 weeks postpartum, pelvic floor symptoms and bother, assessed by the PFDI-20, improved significantly more with pelvic floor physical therapy than standard care.

INTRODUCTION

Women with an obstetric anal sphincter injury are at an increased risk of pelvic floor disorders, including: urinary and anal incontinence; pelvic organ prolapse; and sexual dysfunction.^1,2 While the healing of lacerations that include the anal sphincter may require more time and care, the postpartum care offered to these patients often is no different from patients without advanced lacerations.

In 1948, Kegel first reported the successful treatment of early symptoms of pelvic organ prolapse and stress urinary incontinence using biofeedback-assisted pelvic floor muscle training.³ Pelvic floor muscle training has evolved since 1948, but remains a conservative option for treating pelvic floor disorders. There are some data to suggest that women with anal sphincter injury may benefit from pelvic floor muscle training and other rehabilitative care, commonly referred to as pelvic floor physical therapy (PFPT), during the postpartum period.^4,5 Although the data to support PFPT appear encouraging, there remains significant uncertainty regarding the effectiveness of this treatment.

The aim of this study was to investigate the treatment effect of individualized postpartum PFPT for women with obstetric anal sphincter injury. Our hypothesis was that women with obstetric anal sphincter injuries randomized to a short course of PFPT would have a statistically significant improvement in pelvic floor symptoms and quality-of-life outcomes from 2 weeks to 12 weeks postpartum compared to women who received standard care.

METHODS

We conducted a multicenter, randomized controlled trial at two academic hospitals from December 2014 to July 2017. Women were eligible if they were at least 18 years old, able to read and speak English, and had a vaginal delivery complicated by an obstetric anal sphincter injury (third- or fourth-degree laceration). Women with a history of vaginal delivery after 24 weeks of gestation were not eligible. Potential participants were approached on the postpartum unit. We enrolled and randomized those who gave written informed consent.

We randomly allocated participants to PFPT or standard care in a one-to-one ratio. Block randomization was generated using statistical software and allocation was concealed in sequentially numbered, sealed, opaque envelopes until randomization. One of the study staff enrolled all participants. Both groups received routine postpartum care, which included an appointment with their obstetrician at approximately 6 weeks postpartum. Participants randomized to PFPT also completed up to 12 weekly 1-hour sessions of PFPT, starting within 2 weeks of delivery. All sessions were performed at the same multi-site physical therapy practice by one of several physical therapists trained in pelvic floor and perinatal physical therapy.

Based on the initial evaluation and individual patient goals, a treatment plan was carried out over the sessions. Treatment plans included patient education, manual therapy, therapeutic exercise, neuromuscular re-education, and therapeutic activities and modalities. Sessions consisted of Phase I, from beginning of care until the routine 6-week postpartum obstetrical visit, and Phase II, beginning after that visit and extending through 12 weeks postpartum. Phase I emphasized patient education to facilitate activities of daily living, such as toileting, sitting, lifting and carrying. The physical therapist visually observed the perineum and pelvic floor muscle contraction. Participants were instructed on breathing awareness and exercises to facilitate gentle pelvic floor range of motion and support functional movement. Additional mobility and strengthening exercises for the pelvic floor and pelvic girdle muscles were initiated in Phase I. At the beginning of Phase II, the physical therapist evaluated the pelvic floor, including external and internal evaluation of pelvic floor muscles and laceration scar tissue. PFPT education and manual therapy, including instructions in self-perineal tissue mobilization, were individualized to each participant and progressed as appropriate. Specific education was provided regarding return to sexual activity, return to physical activity, and continuation and progression of functional strength exercises.

All participants completed 5 validated questionnaires at 1 to 2 weeks postpartum, 6 to 8 and 12 to 14 weeks postpartum. We used the Pelvic Floor Distress Inventory-20 (PFDI-20) to assess pelvic floor symptoms and associated bother. The PFDI-20 has 3 subscales: the Pelvic Organ Prolapse Distress Inventory 6 (POPDI-6); the Colorectal-Anal Distress Inventory 8 (CRADI-8) and the Urinary Distress Inventory 6 (UDI-6); higher scores indicate more bothersome symptoms.⁶ We also used the Pelvic Floor Impact Questionnaire-7 (PFIQ-7), which is a condition-specific, quality-of-life measure; higher scores indicate worse functional ability.⁶ The Female Sexual Function Index (FSFI) was used to assess female sexual dysfunction, with a score of 26.55 or lower suggesting sexual dysfunction.^7,8 In addition to the 3 questionnaires, participants also completed the Patient Global Impression of Severity (PGI-S) and Patient Global Impression of Improvement (PGI-I) scales to quantify their perceptions regarding the improvement of their pelvic floor symptoms at 12 weeks postpartum.

The primary outcome was change in the PFDI-20 score from 2 weeks to 12 weeks postpartum. Secondary outcomes were the change in PFIQ-7 and FSFI scores from 2 weeks to 12 weeks postpartum, as well as the PGI-S and PGI-I scores. While there are no data on the PFDI-20 in women with third- and fourth-degree lacerations treated with PFPT, there are data for women with prolapse and incontinence who undergo surgical treatment.⁶ Our sample size calculation was based on these data. We assumed that women randomized to immediate PFPT would experience a reduction of 45 points on the PFDI-20 from baseline to the last PFPT visit at 12 to 14 weeks and that women randomized to standard care would experience a reduction of 20 points over the same time period. The reduction in 45 points on the PFDI-20 was based on mean clinically important difference of the summary score of the PFDI-20 determined by Barber et al.⁹ They found that the mean decline in the summary score for participants undergoing surgery for pelvic floor disorders was 45 points which indicated they were “a little better” on the global rating scale.⁹ We assumed that women assigned to standard care would have a 20 point reduction in symptoms. Assuming a pooled standard deviation of 25 and a two-sided α=0.05, we found that we would have 86% power to detect the hypothesized difference with 20 evaluable participants in each group. Allowing for 20% loss to follow up or withdrawal, we aimed to enroll 25 women per group.

We performed an intention-to-treat analysis. We scored the PFDI-20, PFIQ-7, FSFI, PGI-S, and PGI-I surveys using the standard scoring criteria for each instrument. To evaluate changes in scores on the PFDI-20, PFIQ-7, and FSFI, we calculated the difference in each participant’s score from 2 weeks to 6 to 8 weeks and 12 to 14 weeks postpartum. To evaluate participants’ impressions of whether their pelvic floor symptoms had improved, we compared the proportion of women who reported feeling better on the PGI-I to those who reported feeling the same or worse. Participants also completed the PGI-S to quantify their pelvic floor condition in response to treatment. To address missing outcome data, we used the last observation carried forward. Continuous data are presented as medians and interquartile range (IQR) and were compared using the Wilcoxon rank sum test. Categorical data are presented as frequency (proportion) and were compared using chi-square test or Fisher’s exact test. All tests were two-sided and p-values <0.05 were considered statistically significant. Data were managed in REDCap (Research Electronic Data Capture)¹⁰ and analyzed using SAS 9.4 (SAS Institute Inc, Cary, NC).

The study was registered with the US National Institutes of Health Clinical Trial Registry (NCT02304016) and was approved by the institutional review boards at Mount Auburn Hospital (Cambridge, MA) and Beth Israel Deaconess Medical Center (Boston, MA).

RESULTS

During the study period, we approached 89 eligible women and 50 (56%) women were enrolled and randomized to either the PFPT or the standard care group. The groups were similar with regard to maternal age, parity, gestational age at delivery, and whether they had an operative vaginal delivery or episiotomy (Table 1). In each group, 80.0% of participants had a third-degree laceration and 20.0% had a fourth-degree laceration. Participants randomized to PFPT had a higher body mass index and infants with higher birth weights.

Table 1:

Characteristics at enrollment for all enrolled participants stratified by treatment arm

	Pelvic Floor Physical Therapy n=25	Standard Care n=25
Maternal age	32.0 (29.8, 35.0)	33.4 (31.4, 35.8)
Body mass index (kg/m 2 )	31.9 (27.5, 34.6)	27.4 (25.9, 31.8)
Race
White	19 (76.0)	16 (64.0)
Non-white	6 (24.0)	9 (36.0)
Parity
0	22 (88.0)	23 (92.0)
1	3 (12.0)	2 (8.0)
Assisted Delivery
Vacuum	6 (24.0)	5 (20.0)
Forceps	5 (20.0)	4 (16.0)
Episiotomy mediolateral or midline	4 (16.0)	3 (12.0)
Gestational age at delivery (weeks)	40.0 (38.8, 41.0)	40.2 (39.1, 40.8)
Infant birth weight (g)	3688.5 (3185.0, 4027.5)	3334.5 (2194.0, 3795.0)
Degree of laceration
Third-degree	20 (80.0)	20 (80.0)
Fourth-degree	5 (20.0)	5 (20.0)

Data presented as median (interquartile range) or n (%)

Five participants randomized to PFPT and 1 randomized to standard care withdrew before initiating treatment and did not provide any follow-up data; these 6 participants were excluded from the intent-to-treat analysis. We also excluded 1 participant in the PFPT group who did not provide baseline data. All other participants, 19 in the PFPT group and 24 in the standard care group, were included in the intent-to-treat analysis. Of those randomized to PFPT, 17 initiated treatment and all 17 had outcomes data at 6 to 8 weeks and 15 had outcome data at 12 to 14 weeks (Figure 1). There were no adverse events related to pelvic floor physical therapy.

Figure 1.

CONSORT flow diagram

At 2 weeks postpartum, participants randomized to PFPT had worse median PFDI-20 scores (37.5; IQR: 20.8, 75.0) than the standard care group (18.8; IQR: 8.3, 49.6), indicating they had more pelvic floor symptoms and bother (Table 2). The PFPT group also reported significant improvement in pelvic floor symptoms and associated bother compared with the standard care group, as demonstrated by a decrease of 29.2 (IQR: −58.3, −4.2) compared to 0.0 (IQR: −14.6, 20.8) on the PFDI-20 from 2 weeks to 12 weeks postpartum (p=0.002). Significant differences between the two groups also were observed for the POPDI-6, UDI-6, and CRADI-8 subscales (all p≤0.02). While the PFPT group had a larger decrease on the PFDI-20 than the standard care group, the median PFDI-20 scores were similar in the two groups at 12 weeks (p=0.10; Figure 2). Similar patterns were seen with the POPDI-6, UDI-6, and CRADI-8.

Table 2:

Comparison of score change from 2 weeks to 12 weeks after delivery in pelvic floor physical therapy and standard care groups

	Pelvic Floor Physical Therapy n=19			Standard Care n=24			p
	2 weeks	12 weeks	Change	2 weeks	12 weeks	Change
PFDI-20 Total	37.5 (20.8, 75.0)	8.3 (0.0, 49.6)	−29.2 (−58.3, −4.2)	18.8 (8.3, 49.6)	22.9 (8.3, 43.7)	0.0 (−14.6, 20.8)	0.002
UDI-6	8.3 (0.0, 16.7)	0.0 (0.0, 8.3)	−8.3 (−12.5, 0.0)	0.0 (0.0, 10.4)	0.0 (0.0, 8.3)	0.0 (0.0, 0.0)	0.02
CRADI-8	16.7 (4.2, 33.3	0.0 (0.0, 8.3)	−8.3 (−33.3, −4.2)	8.3 (0.0, 22.9)	10.4 (0.0, 20.8)	0.0 (−12.5, 8.3)	0.01
POPDI-6	12.5 (8.3, 29.2)	0.0 (0.0, 8.3)	−8.3 (−15.0, 0.0)	6.3 (0.0, 16.7)	8.3 (0.0, 20.8)	0.0 (−2.1, 6.2)	0.004
PFIQ-7 Total	57.1 (14.3, 104.8)	9.5 (4.8, 19.0)	−28.6 (−95.2, −4.8)	61.9 (26.2, 130.9)	11.9 (0.0, 50.0)	−38.1 (−116.7, 0.0)	1.0
Bladder/urine	0.0 (0.0, 4.8)	0.0 (0.0, 4.8)	0.0 (−4.8, 0.0)	14.3 (2.4, 44.0)	2.4 (0.0, 11.9)	0.0 (−38.1, 0.0)	0.30
Rectum/bowel	4.8 (0.0, 33.3)	0.0 (0.0, 0.0)	−4.8 (−33.3, 0.0)	16.7 (0.0, 42.9)	0.0 (0.0, 19.0)	−9.5 (−28.6, 0.0)	0.95
Vagina/pelvis	28.6 (9.5, 71.4)	9.5 (0.0, 19.0)	−28.6 (−57.1, 0.0)	33.3 (4.8, 52.4)	4.8 (0.0, 19.0)	−14.3 (−45.3, 0.0)	0.70
FSFI Total	10.0 (7.8, 19.0)	13.6 (7.8, 21.7)	0.0 (−2.1, 5.5)	8.4 (6.4, 9.6)	13.1 (7.9, 19.2)	1.0 (−0.4, 10.7)	0.27
Arousal	1.2 (0.0, 3.3)	2.1 (0.0, 3.6)	0.0 (−1.2, 1.5)	0.0 (0.0, 0.6)	1.5 (0.0, 3.0)	0.0 (0.0, 2.8)	0.18
Desire	4.2 (3.6, 5.4)	3.6 (3.0, 4.8)	0.0 (−1.2, 1.2)	5.4 (4.8, 6.0)	4.8 (3.6, 5.1)	0.0 (−1.2, 0.0)	0.38
Lubrication	3.0 (0.0, 4.2)	3.6 (0.0, 3.9)	0.0 (−1.2, 2.4)	0.0 (0.0, 0.0)	0.8 (0.0, 3.6)	0.0 (0.0, 3.6)	0.27
Orgasm	0.0 (0.0, 3.2)	3.0 (0.0, 4.0)	0.0 (−0.4, 1.6)	0.0 (0.0, 0.0)	0.0 (0.0, 3.2)	0.0 (0.0, 2.8)	0.38
Pain	0.0 (0.0, 4.8)	0.0 (0.0, 3.6)	0.0 (−0.8, 0.0)	0.0 (0.0, 0.0)	0.0 (0.0, 3.0)	0.0 (0.0, 3.0)	0.13
Satisfaction	2.4 (1.6, 4.0)	2.8 (2.0, 4.0)	0.0 (−0.4, 0.8)	2.4 (1.6, 3.2)	2.4 (1.4, 3.0)	0.0 (−0.8, 0.6)	0.50

P-value compares the change in scores between the intervention and standard care groups

Figure 2.

Median Pelvic Floor Distress Inventory-20 (PFDI-20) Score

PFIQ-7 scores indicated that pelvic floor, urinary, and bowel symptoms had a peak impact on daily activities at 2 weeks postpartum and less of an impact at 12 weeks postpartum (Table 2, Figure 3). The median change in PFIQ-7 total score from 2 to 12 weeks postpartum was similar between the two groups (p=1.0; Table 2). As shown in Figure 3, the PFPT and standard care groups had similar PFIQ-7 scores at 2 weeks, 6 weeks and 12 weeks postpartum (all p≥0.45).

Figure 3.

Median Pelvic Floor Impact Questionnaire (PFIQ-7) Score

FSFI scores were low and indicated sexual dysfunction in both the PFPT and standard care groups at 2 and at 12 weeks postpartum. At 12 weeks 12 (63.2%) women in the PFPT group and 15 (62.5%) in the standard care group reported they were not sexually active. Essentially, there was no improvement in the FSFI score from 2 weeks to 12 weeks postpartum, and this was similar for both groups (p=0.27; Table 2).

PGI-S scores at 2 weeks postpartum indicated that 52.6% of the PFPT group and 29.2% of the standard care group felt their condition was normal, while 10.5% of the PFPT group and 8.3% of the standard care group felt that their condition was severe (p=0.34). At 12 weeks postpartum, the proportion who felt that their condition was normal increased to 68.4% and 45.8% in the PFPT and standard care groups, respectively; the difference between the groups at 12 weeks postpartum was not statistically significant (p=0.14; Figure 4).

Figure 4.

Patient Global Impression of Severity

PGI-I scores at 12 weeks postpartum indicated that 72.2% of participants in the PFPT group reported feeling better compared to 50.0% in the control group, though this difference was not statistically significant (p=0.28; Figure 5).

Figure 5.

Patient Global Impression of Improvement

DISCUSSION

We sought to determine if a short course of postpartum PFPT improved pelvic floor symptoms and quality-of-life in the short term among women with obstetric anal sphincter injury. We found that pelvic floor symptoms and bother, as assessed by the PFDI-20, improved significantly in the PFPT group compared to the standard care group. A larger proportion of women in the PFPT group reported on the PGI-I that their pelvic floor symptoms were better at 12 weeks postpartum compared to the standard care group, although this difference was not statistically significant.

Furthermore, not only did we find that the PFDI-20 scores were statistically significant they were also clinically meaningful. The minimal important difference (MCID) is the concept of the smallest change in score that is associated with a clinically meaningful change in quality of life.¹¹ Barber et al. estimated that the MCID of the PFDI-20 for women undergoing surgery for pelvic floor disorders was at least 45 points.⁹ In 2017, Wiegersma et al. determined the MCID of the PFDI-20 for women electing for conservative treatment of prolapse (watchful waiting, PFPT or pessary) and found a change of at least 13.5 points to be clinically significant.¹² In this study, given that women were not undergoing surgery but were proceeding with conservative management, Wiegersma’s estimate in minimal import difference is more applicable. Women in the PFPT group were found to have a decrease of 29.2 points on the PFDI-20, and this difference can be consider clinically relevant.

Our findings contribute to the limited research on the effect of PFPT during the postpartum period. To our knowledge, there is only one other similar randomized trial performed in the U.S.¹³ This study randomized 50 women with obstetric anal sphincter injury to PFPT or expectant management and found no significant difference in improvement in quality-of-life measures between the two groups.¹³ They found that all women had an improvement in quality-of-life measures; however, they only assessed the UDI-6 subscale of the PFDI-20.¹³

Wu and colleagues conducted a meta-analysis of randomized controlled trials, which included the U.S.-based trial described above, comparing PFPT to watchful waiting in women with Stage II or less pelvic organ prolapse within 1 year of delivery.⁵ Three trials performed outside of the U.S. were included in the analysis of pelvic organ prolapse symptoms.^14,15,16 Similar to our study, data from these trials of 609 postpartum women showed a decrease in prolapse symptoms between 6 and 12 months postpartum (risk ratio: 0.48; 95% confidence interval: 0.30–0.76) after PFPT compared with expectant management. However, the authors concluded that there was low-quality evidence that PFPT decreases symptom of prolapse from 6 to 12 months postpartum.⁵

Furthermore, the investigators identified 7 randomized controlled trials, again outside of the U.S., that investigated urinary incontinence, self-reported or diagnosed on exam, during the 12 months postpartum.⁵ Combining 2692 participants, they found a significant decrease in urinary incontinence among participants who underwent PFPT (risk ratio: 0.44; 95% confidence interval: 0.25–0.75), concluding there was moderate-quality evidence for a reduction of urinary incontinence during the postpartum period because of the heterogeneity among the studies.⁵

In regards to sexual function, the two randomized controlled trials that used the FSFI to evaluate postpartum sexual dysfunction found conflicting results.⁵ Thus, the authors concluded there was low-quality evidence that PFPT could improve sexual function.⁵ FSFI scores in our study were low at 2 weeks and continued to be low at the 12 week visit. The initial scores were low because of instructions on pelvic rest for 6 to 8 weeks in the postpartum period. At 12 weeks, greater than half of the participants in both groups were not sexually active. A limitation of the FSFI is that it does not address why a woman is not sexually active and therefore we cannot determine the etiology of the low scores.

The strengths to our study include randomization to minimize confounding and use of validated questionnaires to evaluate symptoms of pelvic organ prolapse, urinary incontinence, anal incontinence, and sexual dysfunction. Furthermore, we only included women who had never had a vaginal delivery after 24 weeks of gestation, given the strong association between vaginal deliveries and pelvic floor disorders.^17,18 Finally, all women underwent PFPT at the same practice by physical therapists trained in PFPT and who used standard treatment regimens.

The limitations of our study include the small sample size and loss to follow-up. It is possible that women in the PFPT group who had fewer pelvic floor symptoms and bother were more likely to withdraw, while women with more symptoms and bother had more motivation to remain in the study and adhere to the assigned PFPT. This could explain the worse baseline PFDI-20 scores in the PFPT group than in the standard care group. In addition, though the PFDI-20 is validated, it has not been validated in patients with obstetric anal sphincter injury. Additionally, we did not obtain objective outcome measures, such as pelvic organ prolapse quantification measurements or anorectal manometry.

PFPT for women with obstetric anal sphincter injury may be beneficial in the postpartum period to improve pelvic floor symptoms and bother. Larger, randomized controlled trials with long-term follow-up are needed to further investigate the role of PFPT in women with obstetric anal sphincter injury.