Periurethral injection of autologous adipose-derived regenerative cells for the treatment of male stress urinary incontinence: Report of three initial cases


Momokazu Gotoh M.D., Ph.D., Department of Urology, Nagoya University Graduate School of Medicine, 65 Tsurumai-cho, Showaw-ku, Nagoya 466-855, Japan. Email:


Objectives:  To report a novel cell therapy using autologous adipose tissue-derived regenerative cells for male stress urinary incontinence caused by urethral sphincteric deficiency, and the outcomes in the initial cases undergoing periurethral injection of adipose tissue-derived regenerative cells.

Methods:  Three patients with moderate stress incontinence after radical prostatectomy and holmium laser enucleation of the prostate were enrolled. Adipose tissue-derived regenerative cells were isolated from the abdominal adipose tissue by using the Celution system. Subsequently, the isolated adipose tissue-derived regenerative cells, and a mixture of adipose tissue-derived regenerative cells and adipose tissue were transurethrally injected into the rhabdosphincter and submucosal space of the urethra, respectively. Short-term outcomes during a 6-month follow up were assessed by a 24-h pad test, a validated patient questionnaire, urethral pressure profile, transrectal ultrasonography and magnetic resonance imaging.

Results:  Urinary incontinence progressively improved after 2 weeks of injection up to 6 months in terms of decreased leakage volume, decreased frequency and amount of incontinence, and improved quality of life. Both maximum urethral closing pressure and functional profile length increased. Magnetic resonance imaging suggested a sustained presence of the injected adipose tissue. Enhanced ultrasonography showed a progressive increase in the blood flow to the injected area. No significant adverse events were observed peri- and postoperatively.

Conclusion:  These preliminary findings suggest that periurethral injection of the autologous adipose tissue-derived regenerative cells is a safe and feasible treatment modality for male stress urinary incontinence.

Abbreviations & Acronyms

adipose-derived regenerative cells


adipose-derived stem cells


colony forming unit-fibroblast


functional profile length


green fluorescent protein


holminium laser enucleation of the prostate


International Consultation on Incontinence Questionnaire-Short Form


magnetic resonance imaging


mesenchymal stem cells


maximum urethral closing pressure

Qmax =

maximum flow rate


quality of life


smooth muscle actin


stress urinary incontinence


Urinary incontinence is a distressing complication of radical prostatectomy and transurethral surgery of the prostate, and is associated with the functional impairment of the external urethral sphincter. Although the incidence of male stress incontinence is not high, a substantial number of patients are known to suffer from long-lasting moderate to severe incontinence after radical prostatectomy.1–3 Although a variety of treatment modalities have been attempted,4 there has been no established treatment for male stress incontinence. Hence, there is an excessive need for developing a new minimally invasive and effective treatment modality for male stress incontinence.

Cell therapy for the regeneration of injured tissues has recently been extensively investigated at an experimental level, and its clinical application in a variety of fields has also been in progress. MSC are multipotent adult stem cells that can proliferate in culture, and are able to differentiate into a variety of mesenchymal cell phenotypes.5–8 Thus far, MSC have mainly been harvested from bone marrow, a tissue source that has many limitations. These include donor-site morbidity in the bone marrow, which limits the amount of marrow that can be obtained,9 MSC represent less than 0.01% of all nucleated bone marrow cells in healthy volunteers,10,11 and an extended culture time is required to obtain therapeutic cell doses of MSC by using the ex vivo cell expansion method.

It has recently been shown that adipose tissue contains multipotent cells that are similar to MSC,12,13 and the abundance of stem cells in the adipose tissue is 100-fold higher than that in the bone marrow. This finding has generated major interest because, unlike bone marrow cells, adipose tissue can be easily and safely harvested in large quantities with minimal morbidity, making it an appealing source for cell therapy. It has been shown that ASC differentiate into several cell types,5,8,14 such as bone, cartilage, fat, nerves, blood vessels and contractile cells with striated muscle cell7 or cardiomyocyte features.7,15 In addition, cultured ASC secrete a variety of angiogenesis-related cytokines, such as hepatocyte growth factor and vascular endothelial growth factor.16

The Celution system (Cytori Therapeutics, San Diego, CA, USA) is commercially available equipment that allows the isolation of stromal, vascular-derived, adipose stem and regenerative cells from human adipose tissue in a short time.17 This instrument allows the isolation of therapeutic doses of autologous ADRC after liposuction, obviating the need for culture. We developed a novel cell therapy for SUI caused by urethral sphincter deficiency; the cell therapy included periurethal injection of autologous ADRC. We report our experience of three initial patients with SUI after prostatectomy undergoing this therapy, focusing on the procedure and short-term outcomes.


The present study was approved by the Ethics Committee of the Nagoya University Graduate School of Medicine, and also by the committee of the Japanese Ministry of Health, Labor and Welfare according to the Guideline on Clinical Research using Human Stem Cells. Written informed consent was obtained from the patients.


In the present study, three patients with SUI after radical prostatectomy and HoLEP were enrolled. The inclusion criteria for the patients with prostate cancer was as follows: persistence of moderate to severe urinary incontinence for more than 2 years after surgery, no evidence of recurrence or metastasis of prostate cancer with undetectable levels of prostate-specific antigen (<0.008 ng/mL), localized prostate cancer of good risk with preoperative prostate-specific antigen of less than 10 ng/mL and a Gleason score of 6 points or less negative surgical margin on pathological examination of the resected prostate specimen. The first patient (case 1) was 77 years-of-age, and underwent radical prostatectomy 4 years earlier with a pathological stage of T2N0M0, and had an undetectable level of prostate-specific antigen at enrolment and moderate SUI without urgency. The second patient (case 2) was 69 years-of-age, and underwent HoLEP for benign prostatic hyperplasia 2 years earlier, and had persisting SUI after surgery. The third patient (case 3) was 75-years-of-age with moderate SUI, and underwent radical prostatectomy 3 years earlier with a pathological stage of T2N0M0, and had undetectable levels of prostate-specific antigen at enrolment.

Harvesting adipose tissue (liposuction)

Under spinal anesthesia, 250 mL of adipose tissue was harvested from the anterior abdominal wall by making two 3-mm incisions. An 18-G Becker cannula with a 50-mL syringe was used as a collecting device; Ringer's Lactate was first infused in the subcutaneous layer then the adipose tissue was harvested. The suctioned adipose tissue contained in the saline was allowed to stand for settling the blood and cellular debris; adipose tissue floated to the top of the mixture.

Isolation of ADRC

ADRC were isolated from the harvested adipose tissue by using the Celution system.18 Briefly, adipose tissue was introduced into the Celution cell-processing device, which automatically and aseptically extracts and concentrates the mononuclear fraction of adipose tissue and removes unwarranted or deleterious cells, and matrix fragments. It required approximately 1 h to process 250 mL of liposuction tissue. The final concentrated cell output collected using the Celution system was counted using a NucleoCounter (Chemometec, Allerød, Denmark), which exclusively detected nucleated cells. By using the Celution system, we could finally obtain a 5 mL solution containing concentrated ADRC.

Periurethral injection of ADRC

Subsequent to liposuction and isolation of ADRC, transurethral endoscopic injection of ADRC was carried out. For periurethral injection of ADRC, two distinct formulations were produced: 1 mL of the isolated ADRC fraction alone was preserved for direct injection, and another 4 mL of the fraction was mixed with intact autologous adipose producing a total of 20 mL of this combined solution.

A 22-Fr rigid nephroscope was used for injecting the processed ADRC cell suspension. Under endoscopic vision, a puncture needle was passed through the nephroscope into the urethra at the region of the external urethral sphincter. The puncture needle with a thickness of 18-G, length of 35 cm and graduated in centimeters was specially ordered. After puncturing the urethra at the region of the external urethral sphincter under endoscopic vision, the ADRC were injected. Initially, a 1-mL solution was injected at a depth of 5 mm into the rhabdosphincter at 5 and 7 o'clock positions. Subsequently, 20 mL of the formulation containing ADRC and adipose tissue was equally injected into the submucosal spaces at 4, 6 and 8 o'clock positions to facilitate complete coaptation of the urethral mucosa by the bulking effect. After the solution was injected, a 6-Fr urethral balloon catheter was placed and removed the following day.

Outcome measures

The amount of incontinence was evaluated by a 24-h pad test. The total daily leakage amount was calculated. The 24-h pad test was consecutively repeated for 4 days for each evaluation period. The subjective symptoms and QOL were evaluated using a validated disease-specific questionnaire – the ICIQ-SF.18,19 In the ICIQ-SF, the therapeutic effects in terms of frequency of urinary incontinence (0–5 point scores), amount of leakage (0–6 point scores) and impact on everyday life (0–10 point scores) were examined, and the total score ranging from 0 to 21 points was calculated. A high score indicated an unfavorable condition. These parameters were assessed at baseline, and repeated 2, 4, 8, 12 and 24 weeks after treatment.

The urethral sphincter function was objectively assessed by measuring the urethral pressure profile using a urodynamic equipment (MMS, Enschede, The Netherlands). MUCP and FPL were measured at baseline, and 2, 12 and 24 weeks after treatment.

The blood flow to the area where ADRC were injected was assessed by contrast-enhanced transrectal ultrasonography by intravenously injecting perflubutane.20 The morphological condition of the injected area was monitored by MRI.


Liposuction from the abdomen was carried out without significant morbidity, and 250 mL of adipose tissues could be harvested. The isolated adipose tissue solution contained 3.3 × 107 ADRC (2.9 × 107 viable cells), 4.0 × 107 ADRC (3.8 × 107 viable cells) and 2.4 × 107 ADRC (2.2 × 107 viable cells) in cases 1, 2 and 3, respectively.

The urethral lumen at the region of the external urethral sphincter remained open, as observed by endoscopy, whereas the urethral lumen completely closed after the periurethral injection (Fig. 1).

Figure 1.

Endoscopic findings during the procedures of periurethral injection of ADRC. The figure shows the findings in case 2. (a) Before injection, the urethral lumen at the region of the external urethral sphincter was open. (b) An 18-G needle punctured the urethra at 4 o'clock of the region of the external urethral sphincter (the arrow shows a puncture needle), (c) After completing the injection, the urethral lumen was closed and complete coaptation of the urethral mucosa was obtained.

Urinary incontinence improved within a few days after injection, deteriorated subsequently and progressively improved thereafter up to 6 months. At 6 months after injection, urinary incontinence improved in terms of leakage volume measured by a 24-h pad test in all cases, and disappeared in case 3 (Table 1). Assessment of subjective symptoms and QOL by the ICIQ-SF showed similar improvement (Table 2). Sphincteric function of the urethra was improved in all cases in terms of increased MUCP and FPL (Table 1).

Table 1.  Clinical outcome for objective findings
 CaseBaseline2 weeks4 weeks8 weeks12 weeks16 weeks24 weeks
24-h pad test (g) during 4 days (mean)1108/134/127/120 (122.3)59/100/80/122 (90.3)80/115/90/58 (85.8)99/110/95/102 (101.5)68/70/88/119 (86.3)98/21/33/104 (64.0)34/45/60/63 (50.5)
230/35/66/67 (49.5)23/54/38/44 (39.8)47/29/22/23 (30.3)43/25/23/27 (29.5)43/25/23/27 (21.8)14/10/13/13 (12.5)12/11/11/12 (11.5)
340/35/25/40 (35.0)39/21/10/28 (24.5)14/0/0/6 (5.0)15/0/0/15 (7.5)9/0/0/6 (3.8)0/6/0/0 (1.5)0/0/0/0 (0)
MUCP (cmH2O)140595253
FPL (mm)120202524
Postvoid residue (mL)125322015
Table 2.  Clinical outcome for symptoms and QOL
 CaseBaseline2 weeks4 weeks12 weeks24 weeks
ICIQ-SF (frequency of leakage)143444
ICIQ-SF (amount of leakage)144433
ICIQ-SF (QOL)166333
ICIQ-SF (total score)11413111010

After the urethral catheter was removed, all the patients could void without significant residual urine. None of the patients complained of voiding symptoms. Uroflowmetry did not show significant voiding dysfunction and an increase in the amount of residual urine (Table 1).

Enhanced ultrasonography showed a sequential increase in the blood flow to the area where ADRC were injected, which was maintained during the entire follow-up period (Fig. 2). MRI showed a bulking effect at the site of adipose tissue injection, which persisted even at 12 weeks after injection (Fig. 3).

Figure 2.

Contrast-enhanced transrectal ultrasonography to assess the blood flow of the periurethral area after ADRC injection. The bladder and urethra was visualized as a sagittal section. The blood flow around the injected area visualized as orange color was progressively increased after the injection of ADRC up to 6 months in case 1, 2 and 3, showing a progressive increase of blood flow at the injection site.

Figure 3.

Magnetic resonance imaging around the urethra before and after the ADRC injection. The urethra in case 1 was visualized as a cross-section on magnetic resonance imaging. (a) Before injection, (b) 4 days after injection and (c) 12 weeks after injection. The arrows demonstrate the adipose tissue injected with ADRC. There was no change in the volume of injected adipose tissue between 4 days and 12 weeks after injection.

No significant adverse event was noted throughout the liposuction and ADRC injection procedures. No severe side-effects, such as pelvic pain, inflammation or de novo urgency, were observed after the operation in all cases during the postoperative follow up.


ADRC have been successfully used in a variety of indications in humans, including the treatment of Crohn's disease-associated fistulas,21 osteogenesis imperfecta,22 and for breast augmentation and reconstruction after partial mastectomy.23 Lin et al. reported the characterization of ADRC isolated with the Celution system in a basic investigation.17 They showed by flow cytometry and CFU-F assays that the isolated cells from the Celution were composed of heterogeneous cell population including ASC, mature and progenitor endothelial cells, vascular smooth muscle cells, CD45+ hematopoietic cells, resident tissue macrophage/monocytes, pericytes, and preadipocytes and so on, containing ASC in 0.6–1.6% of all cell components. To explore the safety and feasibility of ADRC transplantation in patients with myocardial infarction, the first-in-man randomized controlled trial is currently in progress in the Netherlands.15 Our cell therapy is the first attempt to use ADRC for treating SUI.

Before applying this new therapeutic technique to humans, we carried out several animal experiments using rats to confirm the effect of the periurethral injection of ASC on the urethral resistance and sequential changes of the injected rat ASC.24 Cultured rat ASC were injected into the proximal urethra after bilateral transection of the pelvic nerves. Bladder leak point pressure was measured 4 weeks after injection of ASC, Gax-collagen or vehicle. Leak point pressure was significantly higher in the rats undergoing ASC injection as compared with those undergoing injection of collagen or vehicle. Additionally, GFP-expressing cultured ASC obtained from male GFP rats were injected into the urethra of female nude rats. Four weeks after the injection, anti-GFP antibody-positive cells were abundantly stained at the region of ASC injection. Furthermore, 12 weeks after the injection, alpha SMA-positive cells were stained in the merged distribution (70%) with the GFP expressing ASC, suggesting possible differentiation of ASC into smooth muscle cells. These preliminary animal experiments support the present clinical outcomes, such as progressive improvement of sphincteric function and incontinence. Dave et al. also reported the feasibility of ASC use as an improvement in leak point pressure and urethral function of SUI when animals were injected with ASC.25 Recently, a variety of reports have suggested promising feasibility of ADRC and ASC for regenerative treatment of SUI on an experimental level.26,27

Recently, cell therapy using autologous adult muscle-derived stem cells has been developed for treating SUI. A muscle biopsy sample was obtained from the upper arm and cultured to harvest two types of autologous muscle-derived cells: myoblasts and fibroblasts. The muscle-derived stem cells were injected transurethrally into the urethra. Carr et al. reported the outcomes of 1-year follow up of autologous muscle-derived stem cell injection to treat eight women with stress incontinence.28 With a mean follow-up period of 16.5 months, SUI was improved in five patients, with one achieving total continence, and no serious adverse events were noted. Our treatment strategy has an important advantage over the use of muscle-derived stem cells. Because adipose tissue contains abundant multipotent stem cells, as well as key mature cells and progenitor cells, therapeutic levels of regenerative cells can be obtained rapidly using the Celution system. In the present study, a therapeutically relevant number of cells could be isolated from each patient using this system. Unlike other cell therapy strategies, the treatment is all autologous, requires no cell culture and is carried out in the context of a single surgical procedure.

Periurethral injection of autologous fat was previously investigated in female patients with SUI; adipose tissue harvested from the abdomen was transurethrally injected into the submucosal layer under endoscopic vision. Although the injected adipose tissue could sustain a bulking effect, its efficacy was reported to be poor. In a randomized controlled trial that compared the efficacy of fat injection with that of a placebo injection (saline), the improvement rate after fat injection was poor (22%) at 3 months, with no difference to that produced by the placebo.29 These findings suggested that mature adipocytes were unable to survive at the injected site.

In the periurethral injection of ADRC, it has been suspected that a variety of mechanisms are involved in the improvement of the sphincteric function. A similar clinical course in the present three cases implies the involvement of specific factors that can suggest the mechanisms underlying the treatment strategy: urinary incontinence improved within a week after injection, deteriorated subsequently, and progressively improved thereafter up to 24 weeks after the injection. A bulking effect produced by the injected adipose tissue fraction mixed with ADRC is of primary importance. The injected adipose tissue fraction, which was processed to isolate ADRC, contained 30% of lactated Ringer's solution. Absorption of the solution could be responsible for the temporary deterioration in the condition during the initial week. The ASC subpopulation in the ADRC might have contributed to the progressive improvement in sphincteric function, which was shown in terms of increased MUCP and FPL, as well as decreased frequency and amount of urinary incontinence. Persistent bulking effect indicates the survival and growth of the injected adipose tissue, which could also be attributed to the presence of ASC. ASC within the ADRC might differentiate into mature adipose tissue and, possibly, into contractile cells. Previous studies on rats showed that cultured ASC injected into the injured urethra differentiated into contractile cells with smooth muscle cell features.30 Indirect effects of the injected ASC might also be responsible for the improvement. Cultured ASC are known to secrete a large number of angiogenesis-related cytokines.17 In the present study, increased blood flow to the injected area was suggested by ultrasonography. The increased blood flow seemed to be maintained throughout the follow-up period; this increased blood flow could be a result of the angiogenesis effect of the cytokines secreted by the injected ADRC. The increased blood flow might have a positive effect on the regeneration of the injected adipose tissue and impaired sphincteric function.

The present preliminary study showed that periurethral injection of the autologous ADRC is a safe and feasible treatment modality for stress urinary incontinence. We could establish the clinical course and obtained excellent short-time outcomes in the three initial cases undergoing this cell therapy; hence, we intend to increase the number of patients and confirm the long-term outcomes of this treatment modality.


This study was supported by a grant from Nagoya University Hospital.

Conflict of interest

None declared.