Intra-abdominal pressure and pelvic floor health: Should we be thinking about this relationship differently?

Abstract

OBJECTIVES:

Intra-abdominal pressure (IAP) increases during physical activity. Activities with high IAP are often restricted for women because of potential pelvic floor overloading. Researchers categorize high IAP-activities using absolute values (cmH₂0). While essential for descriptive purposes, absolute IAP may not be ideal for individualized exercise recommendations. For oxygen consumption, a well-established measure of fitness, exercise scientists use a percentage of the maximal value observed during exercise to create relative exercise intensity prescriptions for an individual. Relative exercise intensity correlates inversely to the maximal value observed. We explore whether this approach and response pattern extends to IAP observed during exercise.

METHODS:

55 women completed 16 exercises while wearing a vaginal sensor to measure IAP. The highest mean IAP occurred during seated Valsalva/strain (IAP_STRAIN). We calculated relative IAP (%) for each participant by dividing the maximal IAP during each exercise by IAP_STRAIN. We examined relationships between relative IAP and IAP_STRAIN for each activity using Pearson r correlations.

RESULTS:

Mean age was 30.4±9.4 years and body mass index 22.4±2.6 kg/m². For most women, IAP was greater during strain than during exercises. Relative IAPs negatively correlated with IAP_STRAIN. Excluding one exercise due to small sample sizes, r for all others ranged from −0.35 to −0.80, all statistically significant.

CONCLUSIONS:

The relative IAP responses to many exercises exhibit an inverse relationship to the highest IAP values during strain, consistent with other variables measured during exercise. Relative IAP may provide an alternative to absolute IAP in understanding IAP’s effect on pelvic floor health.

Summary Statement:

Relative intra-abdominal pressure (IAP) is an individualized measure that may provide a unique approach to understanding IAP’s effect on pelvic floor health.

INTRODUCTION

Physical activity confers health benefits, yet elevated intra-abdominal pressure (IAP) associated with physical activity may negatively impact pelvic floor health. Activities like running and jumping are recommended to improve bone health but for some women the strain these activities place on the pelvic floor results in stress urinary incontinence^{1, 2}. Though performing high IAP activities is often cited as a potential etiology for pelvic organ prolapse³, a causal pathway for this supposition has not been established. A better understanding of IAP response during exercise may improve future studies examining whether this mechanism increases the risk for pelvic floor disorders (PFDs).

To date, researchers tend to categorize high IAP-activities based upon absolute values^{4, 5}. Certain activities may be encouraged or restricted based on this absolute value of maximal IAP obtained for a sample of women performing the activity⁶. While essential for descriptive purposes, absolute IAP may not be ideal for individualized exercise recommendations. In exercise science, exercise intensity prescriptions are generally based on values relative to each individual’s capacity; response to submaximal activity is described as a percentage relative to the maximal value⁷. For example, if 70% of maximal oxygen consumption (VO₂) is the target for exercise training, a submaximal work rate for an exercise such as walking or cycling is recommended to achieve that relative intensity. The exercise level to achieve the target relative intensity (70% of maximal oxygen consumption) will necessarily vary by an individual’s maximal capacity, in this case, maximal oxygen consumption. A similar approach is used to prescribe safe upper limits for those with health limitations, such as a history of coronary artery disease⁷.

Exploring the relative IAP response during exercise may similarly be useful in understanding individual limits related to pelvic floor symptoms, pelvic floor muscle function and ultimately the development or prevention of PFDs. High absolute IAP values are observed during Valsalva maneuver or maximal abdominal straining⁸ and are generally higher than IAPs obtained during most activities, though variability between individuals is high.^{9 10, 11} Describing absolute IAP during maximal voluntary straining (IAP_STRAIN) may reflect an individual’s upper limits, similar to a maximal fitness value. The relative IAP generated during activities requiring less effort can then be calculated as a percentage of IAP_STRAIN¹². This could provide a context for understanding the effect of IAP on a particular individual.

Thus, the aim of this analysis was to determine whether the IAP response, expressed in relative terms as a percent of IAP_STRAIN, exhibits a pattern similar to measures of physical fitness, such as oxygen consumption. We hypothesized an inverse relationship between the relative IAP during a given activity expressed as a percent of IAP_STRAIN and IAP_STRAIN.

MATERIALS AND METHODS

This secondary analysis of IAP data collected during exercise activities was approved by the University of Utah Institutional Review Board and participants provided written informed consent.¹⁰.

Participants, women aged 18 to 54 years, reported regularly engaging in exercise. Women were excluded who were pregnant or within six months postpartum, had a musculoskeletal injury prohibiting regular physical activity or a positive response on the exercise pre-participation health screening tool, the Physical Activity Readiness Questionnaire¹³. We excluded two women from the original study that did not perform seated strain.

Detailed study procedures are described elsewhere¹⁰. In brief, women completed a laboratory protocol that included exercise and household activities along with seated and lying maximal strain on two occasions. We collected IAP data, which reflect the mean maximal values for each activity, via an intra-vaginal pressure transducer, and downloaded and analyzed data after protocol completion^{10, 14, 15}. For the present study, we used the first of the two data collection sessions. We analyzed data per previously published methods^{16, 17}.

We chose activities for this secondary analysis to reflect traditional exercises and lifting tasks, with specific focus on tasks with progressive levels, such as walking, cycling, lifting, and calisthenics, since increasing activity intensity should translate to increases in IAP (Table 1). Walking, performed at two levels, and running, performed between 8 and 9.7 km/hr, occurred during 30-second intervals on a treadmill (Quinton Q-Stress TM55, Bothell, Washington, USA). Seated and standing cycling (Monark 828E cycle ergometer, Vansbro, Sweden) were conducted at predetermined workloads, to reflect average effort on level grade and harder effort requiring standing, such as climbing a hill, during 30-second intervals. Stand to sit resembled squatting, achieving a 90-degree knee angle in most women. Lifting tasks included two loads in which women lifted a household object from the floor, placed it onto a counter and returned it to the floor, three times. We gave participants basic instructions and a demonstration before each task but did not coach them to follow specific lifting mechanics. Participants performed 8 repetitions of dumbbell seated shoulder press with progressively increasing weights, depending upon ability. Participants jumped in place 10 times. The order of the activity protocol was the same for all participants.

Table 1.

Absolute IAP (cmH₂O) values for seated strain and laboratory activities.

	N	Minimum	Maximum	Mean	Std. Deviation
Seated Strain (Mean of 3 maximal efforts)	55	16.5	207.7	124.9	41.2
Walk 4.8 km/hr 0% grade	55	15.3	36.6	24.7	4.0
Walk 5.6 km/hr 7% grade	55	20.1	56.1	35.7	6.6
Run 8–9.7 km/hr 0% grade	54	32.4	98.7	65.2	13.6
Seated Cycling, 98 Watts	55	4.1	16.5	8.2	2.3
Standing Cycling 147 Watts	54	14.6	66.9	42.8	10.4
Stand to Sit from chair	53	20.6	99.7	37.2	15.7
Lift 13.6 kg floor to counter and back 3 times	54	17.1	62.6	35.0	9.5
Lift 18.2 kg floor to counter and back 3 times	53	13.6	120.0	50.8	19.2
Seated Shoulder Press* 3.6kg	43	4.0	31.9	10.4	5.7
Seated Shoulder Press* 5.5 kg	38	3.9	24.9	11.4	4.8
Seated Shoulder Press* 6.9kg	11	8.1	19.3	12.2	3.6
Seated Shoulder Press* 9.1 kg	8	10.4	36.5	21.8	9.8
Abdominal Curl Ups	55	6.5	82.3	23.2	15.6
Full Sit Ups with feet held	55	13.9	128.5	64.5	24.4
Push Ups from knees	55	4.0	42.5	16.3	6.6
Jumping 10 times	49	25.7	153.8	88.5	24.7

^*8 repetitions of seated shoulder press were done using dumbbells

To measure IAP_STRAIN, participants held their breath, bore down as if having a bowel movement and strained for five seconds for three repetitions each in seated and lying positions. The seated strain demonstrated the highest mean IAP of all the activities assessed. Therefore, we used IAP from seated strain (IAP_STRAIN) as the basis for calculating relative IAP for other activities in the present paper. Observing the highest IAP during seated strain is consistent with the work of others^{8, 12, 18}.

Statistical Analyses

Median (IQR) values for absolute IAP for the original population from which this current study population was derived have been previously published¹⁰. In the current study, we present mean, standard deviation, minimum and maximum values for absolute IAP values (Table 1).

For each participant, we expressed the relative IAP (that is, percentage of IAP_STRAIN) as the IAP for each individual activity divided by the participant’s IAP during seated strain (average of three strains). We used Pearson’s r correlation coefficients to determine the strength and direction of the relationship between relative and IAP_STRAIN. To confirm that the construct of relative IAP was unique from that of the absolute value, we conducted separate correlation analyses between the absolute IAP for each activity and IAP_STRAIN.

Age, BMI, and parity may be related to IAP_STRAIN. To determine whether partial correlation analysis would be more appropriate than bivariate analysis between IAP_STRAIN and each of these variables, we used Pearson r correlation analyses for age and BMI and Spearman’s rho for parity. We observed considerable variability in IAP_STRAIN and 9 women had IAP values that seemed low (<80 cmH₂0). We examined the data and categorized women with IAP_STRAIN < 80 cm H₂0 (n=9) as low and those >/= 80 cm H₂O (n=46) as typical (based on other literature). We compared age, BMI and parity between these groups using one-way ANOVA. One woman in the “low” IAP_STRAIN group had a value ~ 2.5 standard deviations below the mean. Therefore, we conducted separate analyses including and excluding her.

Since this study is the result of a secondary analysis of existing data, power analyses were not conducted a priori. However, a sample size of 46 is adequate to detect correlation of 0.4 with 80% power and a p-value of 0.05¹⁹. The Statistical Package for the Social Sciences (SPSS) (IBM Corp. Released 2013. IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp.) was used to conduct all statistical analyses with alpha level of p<0.05.

RESULTS

The mean (SD) age was 30.4 (9.3) years and mean (SD) body mass index (BMI) 22.4 (2.7) kg/m².

Pearson and Spearman correlation analyses showed a positive association between age and parity (r=.691, p<.001) and age and BMI (r=.318, p=.016). There were no significant relationships between IAP_STRAIN and age (p=.101), parity (p=.311) or BMI (p=.346). Therefore, we did not perform partial correlation analyses.

Age, parity and BMI were not significantly different (p>0.05 for all) between the 9 women with low and 46 women with typical IAP_STRAIN.

Thirteen participants exhibited IAP values for at least one exercise that were higher than the IAP_STRAIN. Nine of these were in the low IAP_STRAIN group. We observed this phenomenon during walking at 4.8 km/hr at 0% grade (N=1), walking at 5.6 km/hr at 7% grade (N=1), running at 8–9.7km/hr (N=3), standing cycling at 900 kg/min (N=2), stand to sit (N=2), lifting 13.6 kg (N=1), lifting 18.2 kg (N=2), full sit-ups (N=4), and jumping (N=9).

The lowest value for IAP_STRAIN (16.49 cmH₂0) was about 2.5 standard deviations below the mean for the group. The Pearson r correlation coefficients including and excluding this participant were not substantively different but increased for some activities with this observation removed. Because this value for IAP_STRAIN seemed implausible, representing a clear outlier, we elected to present results excluding this participant from correlation analyses between relative and IAP_STRAIN.

The mean relative IAP for individual exercises ranged from 8.41% (seated cycling at 98 Watts) to 75.65% (jumping) (Table 2). Pearson r correlation results indicate that most relative IAP values were significantly and negatively correlated with IAP_STRAIN, except for seated shoulder press with 9.1 kg (p=0.557), which only 8 women were able to do.

Table 2.

Relative intra-abdominal pressure, expressed as % of IAP_STRAIN

	N	Minimum	Maximum	Mean (SD)	Pearson correlation
4.8 km/hr 0% grade	54	10.9%	57.2%	21.8 (9.8)%	−.80**
Walk 5.6 km/hr 7% grade	54	15.9%	90.0%	31.7 (15.4)%	−.77**
Run 8–9.7 km/hr 0% grade	53	23.8%	203.2%	56.3 (27.1)%	−.68**
Seated Cycling 98 Watts	54	2.5%	24.1%	7.3 (3.7)%	−.76**
Standing Cycling 147 Watts	53	17.7%	133.2%	37.4 (19.3)%	−.65**
Stand to Sit	52	10.7%	153.2%	32.6 (21.1)%	−.68**
Lift 13.6 kg floor to counter and back	53	10.0%	87.4%	30.1 (13.4)%	−.57**
Lift 18.2 kg floor to counter and back	53	11.5%	113.6%	43.1 (21.3)%	−.53**
Seated Shoulder Press 3.6 kg	42	2.4%	39.7%	9.7 (7.9)%	−.65**
Seated Shoulder Press 5.5 kg	37	4.4%	30.2%	10.0 (5.6)%	−.63**
Seated Shoulder Press 6.9 kg	11	5.3%	24.3%	10.8 (5.1)%	−.67*
Seated Shoulder Press 9.1 kg	8	7.5%	23.6%	17.1 (6.2)%	−.25
Abdominal Curl Ups	54	4.7%	91.5%	20.1 (16.2)%	−.35*
Full Sit Ups w/feet held	54	21.3%	162.4%	56.3 (26.9)%	−.59**
Push Ups from Knees	54	4.8%	56.3%	14.4 (9.2)%	−.57**
Jumping	49	29.2%	165.3%	75.7 (27.4)%	−.65**

^{^}Data shown with outlier removed

^*Significant at p<.05

^**Significant at p<.01

There were no significant correlations between the IAP during most exercises and IAP_STRAIN (data not shown) with the exception of 4 activities [lifting 13.6 kg (r = .338, p=.012), lifting 18.2 kg (r= .274, p=.047), full sits ups (r= .281, p=.038), jumping (r= .467, p=.001)].

DISCUSSION

In this secondary analysis, we confirmed that IAP during seated strain is greater than that generated during most exercises. Consistent with our hypothesis, we found that the relative IAP for many typical exercises does indeed have a negative association with IAP_STRAIN. In most cases, this negative association was of moderate to strong magnitude and statistically significant. These findings support the idea that the relative IAP response to individual, lower-intensity (that is, submaximal) activities exhibit a similar relationship to IAP values observed during voluntary straining as that observed in well-established measures of fitness, such as maximal oxygen consumption.

Our results suggest that designating any absolute IAP value as high may depend upon the upper bounds of a woman’s individual, volitional capacity to generate IAP. To illustrate this, consider results from two of our participants who demonstrated different abilities to generate IAP during strain but with similar IAPs during jumping. The first generates an IAP of 125 cmH₂O during strain while the absolute IAP during jumping is 101 cmH₂O. Her relative IAP during jumping is thus 81%. The second woman can generate more IAP during strain, 176 cmH₂O, and her absolute IAP during jumping is similar, 106 cmH₂O. Her relative IAP during jumping is 60%. Despite similar absolute values of IAP during jumping, the woman with the higher IAP_STRAIN may have experienced lower relative strain from jumping compared to the woman with lower IAP_STRAIN (that is, less ability to generate IAP). This observation is consistent with measures of physical fitness observed in exercise science. One might hypothesize that these findings relate to better fitness in the woman who had higher IAP_STRAIN, but we cannot answer this question with our study as we did not assess physical fitness.

To our knowledge, only one prior study described relative IAP and used Valsalva to characterize maximal IAP¹². In this study of male judo athletes and controls, athletes had higher measures of muscular fitness and maximal IAP than controls. While the absolute IAP during progressive lifting was not different between groups, the relative IAP was significantly lower in athletes; that is, they required lower relative IAP to achieve a similar lifting effort. Similar to our findings, these authors also found a negative correlation between the slope of relative IAP during the various intensities of the lifting task and maximal IAP and the value (r= −0.59, p<.05) was similar to our observations.

The correlations we observed between relative IAP and IAP_STRAIN were negative, though there was variability in correlation strength. We observed the strongest correlations for walking and seated cycling. Other activities are subject to greater task variability between individuals, which may explain lower correlations. Some aspects of task variability that would also result in IAP variability include differences in biomechanics and muscle recruitment and breathing patterns. As we did not coach women during activities, we cannot comment on how each of these factors influenced results. Correlations for the shoulder press progression decreased as weight increased, though this finding is affected by the fewer number of participants who could accomplish this task with higher weight.

Others have demonstrated that IAP increases with increased activity intensity, faster walking speeds and higher loads^{18, 20, 21}. In the present study, relative IAP similarly increased with the activity progressions of walking to running (Figure 1), cycling levels (Figure 2), abdominal curl-ups to sit-ups, and the two levels of lifting (Figure 3). Increasing IAP in response to higher task demand may be due to the role IAP plays in spinal stability and compounded by high acceleration from impact forces²². For example, the greatest mean relative IAP (75.66%) was observed with jumping, an activity requiring spinal stability that also produces high ground reaction forces²³. Seated cycling, without impact forces and posture that supports spinal stability, had the lowest mean relative IAP (8.42%).

Figure 1.

Relative IAP (%) during walking levels and running

Figure 2.

Relative IAP (%) during levels of cycling

Figure 3.

Relative IAP (%) during levels of lifting

Study limitations include the assumption that seated abdominal strain elicits maximal IAP response. Though supported by the literature^{8, 20}, IAP achieved is subject to voluntary maximal exertion, familiarity with the Valsalva maneuver and ability to recruit musculature necessary to perform it correctly. A proper Valsalva maneuver necessitates straining against a closed glottis; this ensures that air is not lost though expiration and enables IAP to rise in the abdominal cavity. Individuals unable to close the airway would diminish the IAP they are able to achieve. Further, some may be unwilling to strain, due to discomfort or fear of urine or gas leakage. The range of IAP during strain reported herein was extremely wide (16.49 to 207.73 cmH₂O), and is unlikely to be indicative of true Valsalva for all, though 84% of our participants generated an IAP > 80 cm H₂O^{9, 24}. Lastly, our findings are limited to women the age and BMI values of our sample and therefore are not necessarily generalizable to those beyond this range.

Considering IAP in absolute and relative values may elicit a change in perspective for clinicians. Rather than considering an absolute IAP value as high, it may prove prudent to reflect IAP for any activity relative to each woman’s volitional capacity before determining whether that exposure is potentially harmful. This is analogous to the change that has occurred over time in exercise recommendations during pregnancy. In 1985, ACOG recommended that pregnant women should not exceed a heart rate of 140 beats per minute during exercise²⁵, an example of an absolute value considered appropriate for all pregnant women. However, for a fit woman, achieving a heart rate greater than this cut-point is not taxing or harmful, whereas for a pregnant non-exerciser, exceeding the same cut-point may result in adverse effects. Thus, ACOG has since changed this recommendation, no longer considering heart rate in its guideline. Some countries’ guidelines maintain heart rate cut-points and others provide target heart rates of 60% to 80% relative to maximal aerobic capacity^{26, 27}.

Considering the unique IAP response each woman experiences with a given task, relative to her volitional capacity, may broaden our understanding of the beneficial and detrimental roles of IAP response to exercise, thus ultimately enabling clinicians to take a more holistic and individual approach to physical activity prescription as it relates to pelvic health.