Introduction

In pregnancy, as a result of anatomical changes associated with growing uterus and hormonal adjustments with increased estrogen, progesterone and relaxin, and changes in collagen levels [1], there is a biomechanical change in the function of the lower urinary tract, which may be manifested by symptoms of nocturia, incontinence, urgency, and frequency [2].

Urinary incontinence (UI) is common in pregnant women and seriously affects personal and social life, self-esteem, sexuality, and quality of life of affected individuals [3]. The three main types are urgency urinary incontinence (UUI), urine leak is associated with a sudden compelling desire to void, caused by overactive bladder, stress urinary incontinence (SUI), in which urine leaks result from physical exertion, commonly caused by weak urethral muscle and support, and mixed urinary incontinence (MUI) with the combination of UUI and SUI [4].

The prevalence of UI increases later in pregnancy, being SUI the most frequent. The risk factors that contribute to a higher prevalence of urinary incontinence are: maternal weight gain, advanced gestational age, parity, vaginal delivery, urinary tract infections, and constipation [5].

The prevalence of UI in women with type 1 diabetes was greater than neuropathy, retinopathy and nephropathy [6] and the clinical condition is able to alter urination patterns and contribute to incontinence through glycosuria, microvascular damage, and neuropathy [7]. In these women, raising awareness about glycemic control levels is important since there is an association with the risk of developing UI [8].

The literature does not consistently show the impact of type 1 diabetes mellitus during pregnancy on maternal urinary complaints. The aim of this study was to describe the prevalence of UI reports in pregnant women with type 1 diabetes and the impact of this problem on their quality of life.

Methods

Study Design

This cross-sectional case control study was conducted at the Ribeirão Preto Medical School (FMRP-USP), from February 2017 to January 2019. Data collection of each pregnant woman was done in a single day, lasting about 1 h for each interview and all were performed in the morning, in order to limit the effects of hormonal variations during the day [9].

This research study was approved by the Research Ethics Committee of the Clinical Hospital of the Ribeirão Preto Medical School at University of São Paulo (CEP-HCFMRP-USP), Certificate of Presentation for Ethical Assessment number 61052016.3.0000.5440, and all women who accepted to participate in the research signed the Informed Consent Form.

The sample was calculated based on a previous study using the ICIQ-SF score with results of 2.66 ± 4.1 (mean ± SD) with a statistical power of 80% and an alpha error of 0.05 [10]. A total of 44 women in the third trimester of pregnancy was recruited (25 weeks or longer), however, four gave up before finishing the research, after starting the evaluations, because they would be no longer available. Therefore, 40 pregnant women were allocated into two groups, the type 1 diabetes mellitus group (1DMG) diagnosed for at least 5 years (n = 20), and the control group (CG), pregnant women experiencing a healthy uncomplicated pregnancy (n = 20) at a age range between 18 and 40 years old, in good physical and mental health.

Exclusion criteria were: infectious processes; multiple pregnancy; haemodynamically significant disease; restrictive lung disease; cervical incompetence/cerclage; multiple gestation, risk of preterm birth; persistent bleeding in the second or third trimester; placenta previa; preterm labor; rupture of membranes; pre-eclampsia and uncontrolled hypertension; and cognitive changes.

Experimental Procedure

Standard Interview

Initially, the participants answered standard interview (collecting information about age, gestational age, number of episiotomies, number of pregnancies, number of normal and cesarean deliveries and abortions).

Physical Evaluation

The physical evaluation consisted of body mass index (BMI) and uterine height. The glycated hemoglobin value (HbA1c) of the pregnant women belonging to the type 1 diabetes mellitus group was obtained from the records of the Clinical Hospital of the Ribeirão Preto Medical School at University of São Paulo.

International Consultation on Incontinence Questionnaire - Short Form

One way to assess the existence and impact of urinary incontinence is through the International Consultation on Incontinence Questionnaire - Short Form (ICIQ-SF) validated in 2004 for Portuguese, with Cronbach’s alpha 0.88 [11]. It is a self-administered and easily understable tool.

It has shown that a significant correlation exists between this questionnaire, the 24-hr pad test [12, 13] and Urodinamics parameters [11, 14,15,16].

ICIQ comprises four items, three are scored and summed to yield the total score (score 0–21). These questions assess the frequency of leakage (score 0–5), the amount of leakage (score 0–6), and the impact of incontinence on the quality of life (score 0–10). The fourth unscored question is self-diagnostic about the perceived causes of incontinence [15]. We analyzed ICIQ of two isolated forms, the first one was through the sum of the scores of questions 3, 4 and 5 (ICIQ score) and, the second, through the answers of question 6, which reveals in which situation the episodes of urinary incontinence happen [15, 16].

Pregnancy Physical Activity Questionnaire

The participants also answered the Pregnancy Physical Activity Questionnaire (PPAQ), translated to Portuguese, Cronbach’s alpha 0.7 [17].

PPAQ is a self-administered semiquantitative questionnaire to report the level of physical activity done during pregnancy. It takes into account the time spent in 32 activities: household/caregiving (13 activities), occupational (5 activities), sports/exercise (8 activities), transportation (3 activities), and inactivity (3 activities). For each question, the participant must indicate the average time spent, during the current trimester, in a week or day in the respective activity. At the end of PPAQ, there is an open-ended section allowing the respondent to add not listed activities. For calculating the level of physical activity, this questionnaire takes into account the metabolic equivalent of task (MET) of each activity [18].

Weight of the Newborn

The presence of fetal macrosomia is common in diabetic pregnant women [19]. To find out if this change occurred in our participants, in an attempt to standardize the samples, after birth, we contacted the participants through the telephone number given at the moment of the standard anamnesis, to record the weight of the newborn at the time of delivery.

Statistical Analysis

Initially, the data normality assumptions were verified through the Shapiro-Wilk statistical test (alpha = 5%) and homoscedasticity among the variances using the Levene statistical test. In the presence of normality, the groups were compared by Student’s t test at 5% probability. In contrast, in the absence of normality the groups were compared by the U-Mann Whitney test also at a 5% probability. The analyses were processed in SPSS version 14.0, SPSS Inc., USA.

Correlation tests for the time of diagnosis of type 1 diabetes and HbA1c in relation to the variables of the International Consultation on Incontinence Questionnaire Short Form (ICIQ-SF) and the weight of the newborn were performed. We also correlated ICIQ-SF score with the number of previous episiotomies, gestations, vaginal and cesarean deliveries, BMI, newborn weight, uterine height, PPAQ and gestational week for both groups. For the parametric variables, the Pearson’s test was used and, for the nonparametric ones, the Spearman’s test.

We did a comparative descriptive analysis for the urinary incontinence types, reported in each group in ICIQ-SF.

Results

The anthropometric characteristics evaluated showed similar values between the two groups, the mean gestational age in weeks being 31.15 (SD 3.91) for the CG and 30.35 (SD 3.87) for the 1DMG; mean age was 31 (SD 4.21) years for CG and 25 (SD 0.92) years for 1DMG, body mass index was 27.40 (SD 3.00) kg/cm2 and 26 0.71 (SD 5.84) kg/cm2 for the 1DMG and the average weight of the newborn was 3.300 (SD 0.39) kg and 3.430 (SD 0.50) kg.

In 1DMG, the glycated hemoglobin (HbA1c) was examimed and the mean and standard deviation are 6.87 ± 1.57%.

Table 1 shows the comparison between the groups showing that the numbers of episiotomies and abortions, and ICIQ value is higher in 1DMG compared to CG.

Table 1 Comparison between the groups about the numbers of episiotomy, pregnancies, vaginal delivery, cesarean delivery, abortions, ICIQ total score, uterus fundal height and PPAQ total score of the control group and type 1 diabetic mellitus group
Full size table

The correlation tests for the time of diagnosis of type 1 diabetes and HbA1c in relation to the variables of ICIQ and the weight of the newborn do not have significant correlations (p < 0.05). In addition, when we correlated ICIQ, and total score with the number of previous episiotomies, gestations, vaginal and cesarean deliveries, BMI, newborn weight, uterine height, physical activity questionnaire and gestational week for both groups we did not find significant correlations either (p < 0.05).

We analyzed descriptively the groups answers to ICIQ questions 3, 4 and 5 and the types of urinary incontinence discovered by the last ICIQ-SF question and the Results are shown in Tables 2 and 3, respectively.

Table 2 ICIQ-SF Questions 3, 4 and 5 - comparison between the control group and type 1 diabetic mellitus group
Full size table
Table 3 ICIQ-SF Question 6 - comparison between the control group and type 1 diabetic mellitus group
Full size table

The answers 6.2 (leaks before you can get to the toilet), 6.4 (leaks when you are asleep), and 6.6 (leaks when you have finished urinating and are dressed) were an indicator of UUI. Answers 6.3 (leaks when you cough or sneeze) and 6.5 (leaks when you are physically active/exercising) were an indicator of SUI, combination of these answers was an indicator of MUI. Answers 6.7 (leaks for no obvious reason) and 6.8 (leaks all the time) were assigned to other causes of UI [15, 16]. The data were organized and are shown in Fig. 1.

Fig. 1
figure 1

CG (control group); 1DMG (type 1 diabetic group); SUI (stress urinary incontinence); UUI (urgency urinary incontinence); MUI (mixed urinary incontinence); Normal (pregnant without urinary incontinence). The figure shows how many cases of each type of urinary incontinence, followed by the percentage

Full size image

Discussion

The urinary changes caused by pregnancy can be exacerbated by the presence of type 1 diabetes mellitus. The present study evaluated the possible influence of the disease on urinary continence described by the International Consultation on Incontinence Questionnaire Short Form (ICIQ-SF) of pregnant women in the last trimester, which is probably the most affected period.

The limitations of this study rely on the fact that this is a comparative cross-sectional study between two groups (n = 20), acquired after sample calculation although it indicates some differences in the postural control of diabetic pregnant women. However more studies must be carried out to deepen the knowledge about the characteristics of diabetic women during the gestational period.

The findings of this study show the prevalence of urinary incontinence in women with and without type 1 diabetes mellitus. Among pregnant women without diabetes, four participants have stress urinary incontinence, it can be explained by the fact that, during pregnancy, physiological changes are one of the main risk factors for SUI development, the increase in cases of incontinence can be observed between the 1st and 3rd trimester [1, 20]. The weight of the fetus leaning over the pelvic floor muscle and the increasing pressure of the growing uterus on pelvic structures, together with reduced sphincteric, strength and supportive function of pelvic muscle floor caused by hormonal changes, with relaxin and collagen leves and increased progesterone, what may impair the functionality of the lower urinary tract [1].

The relaxin hormone is produced by the corpus luteum and, after 3 months of pregnancy, the placenta takes over the production [21]. It is known that relaxin causes depolymerisation of the inter and intramolecular collagen bonds, loosening the pubourethral ligaments collagen, what explains the stress incontinence [20]. Although our study did not provide data on urinary urgency for CG, it is known that it may happen and is explained by laxity in all the ligaments that suspend the anterior vaginal wall, not supporting the stretch receptors at bladder base, activating the micturition reflex prematurely [20].

Diabetes can interfere not only with the function of the lower urinary tract [22], but also with the development and quality of labor [23]. We reported an increase in the rate of previous episiotomies (p = 0.04) and abortions (p = 0.004) in diabetic women (1DMG) compared with the control group (CG), what agrees with the findings that these women experience a higher risk of episiotomy [24] and abortion [25]. The increase in the rate of episiotomies was not correlated with the prevalence of UI assessed by ICIQ (p = 0.163; r = 0.323) as observed in the literature [26].

In addition to changes in the gestation itself, diabetic pregnant women also had an increase in the total value obtained in ICIQ with score of items 3, 4 and 5 (p = 0.026). And, regarding the analysis of ICIQ question 6, it was possible to observe a discrete increase in the number of participants with stress urinary incontinence (increase of 1 case) in 1DMG when compared with CG, also the presence of urgency urinary incontinence (3 cases) and mixed urinary incontinence (2 cases) in 1DMG, which did not occur in CG.

Urinary incontinence is common in women with type 1 diabetes. Some studies suggest it may be associated with higher HbA1c levels [27, 28], but our results show no statistical relevance. The biology of diabetes associated bladder complications can be due to an alteration in the detrusor smooth muscle, neuronal dysfunction and urothelial dysfunction. Specifically it is hypothesized that microvascular complications might damage the innervation of the bladder or alter bladder muscle function [29]. Recent research in diabetic rats has shown the interference in pelvic floor muscle with steady decrease in the proportion of fast to slow fibers, and a colocalization of fast and slow fibers, fibrosis, atrophy, thinning, mitochondria accumulation, increased area of blood vessels and lipid droplets, and associated glycogen granules [30, 31].

The present study shows that in pregnant women with type 1 diabetes mellitus in the third trimester of gestation seem to be associated with an increase score in International Consultation on Incontinence Questionnaire - Short Form (ICIQ-SF), showing a higher impact of urinary incontinence in their lifes.