Of the 113 patients, prior to THA, UI was prevalent among 59 (52%) patients (incontinent group) and was absent among 54 (48%) patients (continent group).

There was no significant difference in the number of births, age, body mass index (BMI), intraoperative bleeding, or surgery time.

Concerning ROM, there was no significant difference in flexion, abduction, adduction, external rotation, or internal rotation. However, there was a significant difference in extension (P < 0.05).

In the incontinent group, the symptoms of UI improved in 50 (85%) patients, remained unchanged in 5 (8%) and worsened in 4 (7%) after THA.

In the continent group, the symptoms of UI remained unapparent in 52 (96%) patients, and worsened in 2 (4%) (Table 1).

Among the 113 patients, the number of patients whose UI improved was 50 (group A), and whose UI remained unchanged or worsened was 63 (group B). Characteristics of the 2 groups are listed in Table 2. There were no significant differences in age, height, number of births, unilateral or bilateral THA. There were significant differences in weight and BMI (P < 0.05, P < 0.01 respectively). In response to the question, “when does urine leak?” for the patients who had UI symptoms before surgery, 44 (75%) replied “when coughing or sneezing” the most frequently (Figure 1).

-Group A-

Flexion significantly improved from 85.5 ± 20.1 before THA to 91.6 ± 9.5 at 2 weeks after (P < 0.05).

Extension significantly improved from −3.4 ± 7.8 before to 0.6 ± 5.4 at 2 weeks after (P < 0.01).

Internal rotation significantly improved from 13.2 ± 12.9 before to 27.8 ± 11.9 at 2 weeks after (P < 0.01).

There was no significant difference in abduction, adduction, or external rotation (Table 3).

-Group B-

Extension significantly improved from −2.5 ± 7.9 before to 2.8 ± 6.2 at 2 weeks after (P < 0.01).

Abduction significantly improved from 13.3 ± 0.9 before to 15.9 ± 4.8 at 2 weeks after (P < 0.01).

Internal rotation significantly improved from 17.2 ± 15.9 before to 26.4 ± 11.3 at 2 weeks after (P < 0.01).

There was no significant difference in flexion, adduction, or external rotation (Table 4).

Improvement in ROM was compared between the 2 groups (Figure 2). Flexion, extension and internal rotation were significantly improved at 2 weeks after THA in group A. In group B, extension, abduction and internal rotation were significantly improved at 2 weeks after THA. The difference between unilateral and bilateral arthroplasty was compared, and there were no significant differences in terms ROM. We determined improvement rate by dividing the ROM

before THA by the ROM after 2 weeks THA. Using the improvement rate of ROM, statistically there was a significant difference between 2 groups, only in internal rotation (P < 0.05). Improvement rate of internal rotation was significantly higher in group A.

The patient was a 55-year old woman who was suffering from right hip pain because of her osteoarthritis of the hip. According to her questionnaire before THA, she leaked a moderate amount of urine two or three times a week. And her leaking urine interfered with her life by level 4. Her Points of ICIQ-SF was 10 before THA. 2 weeks after the THA, she never leaked urine, and it interfered with her life by revel 1. Her point of ICIQ-SF was 1, 2 weeks after the THA (Figure 3).

Kikuchi et al. [9] investigated the prevalence of UI using ICIQ-SF in 676 Japanese men and women aged > 70 years who were living in the Tsurugaya area of Sendai, one of the major cities in the Tohoku area of Japan. In the study, the prevalence of UI was 25% (34% in women and 16% in men), and the shown high physical activity level was independently related to a lower self-reported prevalence of UI in a community-dwelling elderly population aged > 70 years.

Tamaki et al. [7] reported the prevalence rate of UI in patients with hip osteoarthritis before THA was 43%, using ICIQ-SF.

Baba et al. [8] reported the prevalence rate of UI in patients with hip osteoarthritis before THA was 47% using ICIQ-SF.

Okumura et al. [3] reported the prevalence of UI in patients with hip osteoarthritis before THA was 67% using a core lower urinary tract symptom score (CLSS).

In our study, the prevalence of UI in patients with hip osteoarthritis before THA was 52%.

In the case of hip joint disorders, physical activity levels might be lower, and the prevalence of UI in patients with hip dysfunction may be higher than in people without hip dysfunction.

Hannestad et al. [10] reported half of the incontinent women were experiencing symptoms of stress incontinence (SUI) alone, symptoms of urge incontinence (UUI) alone affected only one in ten, while mixed incontinence (MUI) was

reported by one in three in 6501 of UI patients. In our study, stress incontinence was the most frequent. (Figure 1) Bo K reported that pelvic floor muscle training is effective in the treatment of female stress incontinence (SUI) and mixed urinary incontinence and therefore, it is recommended as a first-line therapy.

The stress continence system includes the sphincteric closure mechanism consisting of urethral striated muscle, urethral smooth muscle, and vascular elements and the remainder of the bladder support system consisting of the anterior vagina, endopelvic fascia, tendinous arch of levator ani, and bony pelvis [11] . The consequence of levator ani muscle damage might depend on which or how many elements have been damaged [12] . If the tendinous arch of levator ani falls, the ipsilateral side of the vagina falls, carrying with it the bladder and the urethra, and thus contributing to urinary incontinence. The tendinous arch of levator ani is at the bottom of the fascia obturator.

Tamaki et al. focused on the obturator internus muscle, which originates from the pelvic floor and has a close relationship with the levetorani muscle [7] . Okumura et al reported that THA is performed to improve hip-joint function and recovery of levator ani function that is connected to the obturator internus [3] .

Baba et al. [8] reported that since the short external rotator may have been atrophied due to hip joint dysfunction before surgery, support of the pelvic organs and UI may be improved if the strength of this muscle group recovers. And, symptoms of UI were significantly improved in the anterior approach group without dissecting the short external rotators, and aggravated in the posterior approach group [8] .

The superior gemellus, inferior gemellus, and obturator internus muscle have been regarded as a single muscle unit and called the “rotator triceps muscle” [13] . The gemelli converge onto the tendon of insertion of the obturator internus muscle and because the three muscles terminate at the same common bony attachment, the medial aspect of the greater trochanter [13] . Their functions are also common; they act synergistically to rotate the pelvis and/or femur [13] . The obturator internus muscle arise over a wide area of the lesser pelvis, converges to the lesser sciatic notch, changes its course, and is inserted into the medial aspect of the greater trochanter [13] .

In our institution, all the arthroplasty procedures employ the anterolateral muscle-sparing approach in the 60-degree half lateral position. At the time of stem insertion, we may cut the conjoint tendon as little as possible if it is hard to insert the stem.

In our study, UI improved in 50 (85%) patients. In other studies, Tamaki et al. [7] reported the symptoms of UI ameliorated in 52 (64%) patients which was done by the muscle-sparing direct anterior approach using ICIQ-SF. Okumura et al. [3] reported in total, patients were better than improved (72%). The rate of cured and improved patients was 76%, for stress UI, 100% mixed UI, and 50% urge UI. Baba et al. reported UI improved after THA in 8 patients (22%), slightly improved in one (2.8%) in the anterior approach group using the distal part of the Smith-Petersen approach. In the posterior approach group, by dissecting the short external rotators, UI improved after THA in one (2.5%). Patients with hip disorder have limited range of motion, and those muscles around hip reduce tension. In our study, improvement rate of internal rotation of group A (improvement of UI group) was high. Think about why UI improved after THA, and the internal rotation movement became possible after that, external rotators including the obturator internus are able to stretch, and preserve the original musculature. As a result, the obturator internus can exert tension. The obturator internus is closely related to the levator ani. In addition, thinking of the approach of THA which preserves the obturator internus, it could be advantageous for the improvement of UI.

Pelvic floor dysfunction is a common problem that leads to significant suffering among women [14] . Pelvic floor muscle training is effective in the treatment of female stress (SUI) and mixed urinary incontinence and, therefore, it is recommended as a first-line therapy [11] . Hip joint disorders from osteoarthritis and UI both greatly reduce the QOL. And both ages of onset are the same for middle age and older [7] . Tamaki et al. hypothesizes that a loose pelvic floor could be improved by THA [7] . In our study, 85% of the THA patients’ UI improved just 2 weeks after THA. That suggests THA might have the same effect as pelvic floor muscle training and might be more immediately effective than muscle training.