The use of autologous skeletal muscle progenitor cells for adjunctive treatment of presumptive urethral sphincter mechanism incompetence in female dogs

2.1 Animals

Spayed female dogs that had urinary incontinence at rest and been diagnosed with USMI but failed standard medical management were recruited for this prospective, single-armed clinical trial. The diagnosis of USMI was confirmed by verifying clinical signs that were consistent with USMI (ie, incontinence at rest and not while walking), failure to document another cause of incontinence through standard clinical evaluation (complete history and physical examination, serum biochemistry panel, CBC, urinalysis, urine culture, abdominal ultrasound examination and vaginocystourethroscopy). Failed medical management was defined as either continued incontinence despite drug administration (i.e., phenylpropanolamine [PPA] or an estrogen (e.g., estriol, diethylstilbestrol [DES]), administered alone or in combination at doses within the recommended ranges published in companion animal formularies) or the presence of unacceptable adverse effects of the medication (e.g., anxious behavior, aggression).^4-6 Dogs were excluded if they had a congenital anatomic cause for their incontinence (e.g., ectopic ureter, congenital vaginal abnormality), a pattern of incontinence that was not compatible with a primary diagnosis of USMI (i.e., if the leakage of urine occurred primarily while walking or after voiding), or unresolved urinary tract infection. The study was approved by the Institutional Animal Care and Use committee of North Carolina State University College of Veterinary Medicine; all participating owners gave informed consent.

2.2 Assessment of continence

Continence was assessed at baseline and 3, 6, 12, and 24 months after skMPC injection using a urinary incontinence questionnaire (Figure S1) and urethral pressure profilometry (UPP). Based on the answers the owners provided to the urinary incontinence questionnaire, dogs were assigned a continence score of 1 to 5, with 1 being a dog that is never continent and 5 being a dog that is always continent. Answers to the questionnaire also were used to determine if dogs were experiencing any outward signs that might suggest an adverse reaction to the skMPC injection and to record concurrent medications used to improve continence.

Urethral pressure profilometry was used to measure MUCP as previously described.^{20, 21} For the UPP procedure, anesthesia was induced in each dog by administration of propofol (4-6 mg/kg IV). Immediately afterward, dogs were intubated, connected to a semi-closed circle system with sevoflurane as the inhalant, and mechanically ventilated. A multi-gas analyzer (Gas Module SE, Mindray, Mahway, New Jersey) was connected to the endotracheal tube to measure the end-tidal sevoflurane concentration. The UPP was initiated after at least 30 min had elapsed since injection of propofol and 10 min had elapsed since decreasing the end-tidal sevoflurane concentration to 2.0%.

During the intervening period between anesthesia induction and UPP initiation, the rectum was manually evacuated and an abdominal sensor catheter (TDOC-7FA, T-DOC Co, Wilmington, Delaware) was placed through the anus and into the colon to a point approximately 15 cm from the anus. The vulva and surrounding area were clipped and prepared using betadine solution and saline. Using sterile technique, a dual sensor catheter (TDOC-7FD, T-DOC Co, Wilmington, Delaware) was passed through the urethra to the level of the bladder. The system and catheters were prepared following manufacturer instructions that accompanied the urodynamic measurement system (Goby KT, Laborie Medical Technologies, Mississauga, Ontario). The bladder was infused with warmed saline until urine leaked from the patient or, if leakage did not occur, until a volume of 10 mL/kg was achieved. The catheter was clamped to the mechanical puller arm and the puller speed was set to 1 mm/s. The tracing was observed as the recorded urethral pressure increased and then decreased to baseline. The pull was stopped. The procedure was repeated 3 times. Maximal and baseline urethral pressures were recorded. The baseline urethral pressure was subtracted from the maximal urethral pressure to derive the MUCP.

2.3 Isolation of skMPCs

After baseline UPP, a triceps muscle biopsy was obtained for harvesting of the skMPC. The plane of anesthesia was increased after UPP to allow for muscle biopsy; protocols varied, as determined by the attending anesthesiologists. Each patient was placed in lateral recumbency, and aseptic preparation of the triceps biopsy site was performed. A 5 cm incision was made through the skin and SC fat using a #10 scalpel blade. Gelpi retractors were placed, exposing the fascia of the lateral head of the triceps muscle. An elliptical incision was made in the triceps using the scalpel and an approximately 1 cm³ sample of triceps muscle was removed using Metzenbaum scissors. The biopsy sample was placed in a sterile moist gauze within a sterile specimen cup and submitted for skMPC isolation. The triceps muscle was closed with 2-3 cruciate sutures using 3-0 to 4-0 polydioxanone suture (PDS). The SC tissue was closed with a simple continuous pattern using 3-0 to 4-0 PDS. The subcuticular layer was closed in a continuous horizontal mattress pattern with 4-0 PDS or Monocryl (Monocryl, Ethicon, Cornelia, Georgia). The skin incision was opposed with tissue adhesive.

Biopsy specimens were minced using scalpels and connective tissue was removed. Minced muscle tissue was digested in 0.2% type I collagenase. After digestion, tissue was plated on Matrigel-coated plates to allow for the expansion of skMPC. As cells expanded, they were passaged 2 to 3 times at 50% to 70% confluence. Cell viability was assessed by visual inspection before harvest for injection; cells were assessed to be viable if they were adherent to the surface of the tissue culture plate, were compact and round to oblong and refracted light when viewed by phase contrast microscopy.²² Once cell numbers reached 100 million, cells were harvested and resuspended 0.6 mL of phenol-free Dulbecco's Modified Eagle Medium (DMEM) in preparation for injection.

2.4 Injection of skMPCs

Once adequate cell numbers were reached, the patient was anesthetized using an anesthetic protocol that was determined by the attending anesthesiologists. Each patient was placed in dorsal recumbency, and the surgical site was aseptically prepared. A 10 to 15 cm ventral midline incision was made using a #10 scalpel blade starting in the caudal abdomen and extending to the pubis. The SC tissues were bluntly dissected, and hemostasis was achieved using electrocautery. A 10 cm incision was made in the caudal linea alba. The bladder was exteriorized, and a single stay suture was placed in the apex of the bladder. The skin, SC tissues, and linea were retracted using Weitlaner or Gelpi retractors just cranial to the pubis. The proximal urethra was located, and fat was cleared from the ventral and lateral aspects of the urethra using blunt dissection with right-angle forceps and cautery.

In 3 separate syringes, approximately 0.2 mL of cell suspension was mixed with approximately 0.13 mL of the dog's own plasma immediately before injection. The plasma was added so as to form autologous platelet gel that could provide a nurturing scaffolding for the stem cells.²³ Each aliquot was injected via a new 25-gauge needle into the urethral muscularis evenly divided among 3 evenly spaced sites circumferentially around the proximal urethra. One milliliter of material (0.6 mL of cell suspension and 0.4 mL plasma) was injected. Cystourethroscopy was used to visualize the internal lumen of the urethra and verify that the needle did not penetrate through the urethral wall into the lumen. The 3 injection sites were spaced approximately 120° apart, with the first injection on the ventral aspect of the urethra. A bleb of suspension could be visualized within the wall at each site both externally and internally (via the endoscope). The stay suture was removed from the apex of the bladder.

The linea was closed with 0 to 2-0 PDS in a simple continuous pattern. The linea and SC tissues were blocked with 2 mg/kg of 0.25% bupivacaine. The SC tissue was closed using a simple continuous pattern with 3-0 PDS. The subcuticular layer was closed in a continuous horizontal mattress pattern with 3-0 to 4-0 PDS or Monocryl (Ethicon, Inc., J&J Surgical Technologies, Bridgewater, New Jersey, USA). The skin was opposed using staples or tissue adhesive.

After recovery from general anesthesia, dogs were hospitalized overnight and discharged the next day. If there were no contraindications, carprofen (4.4 mg/kg IV, once) was administered at recovery and then continued PO (2.2 mg/kg q24h) for 3 to 5 days starting 24 h after the IV dose. Antibiotics were administered only if previously prescribed for the treatment of urinary tract infection. Between anesthetic recovery and discharge, the incision was evaluated, and pain scored q4-6h. A cold compress was applied to the incision q6h. Narcotic analgesia was provided as needed q4-6h if the pain score was ≥2 out of 4.

Medications used to maintain continence were maintained in the first 3 months post-injection to allow for tissue regeneration and remodeling and were discontinued 3 months post-injection. After that time point, medications were administered to improve continence as needed as determined by the attending clinician.

2.5 Statistical analysis

An omnibus test was conducted in the form of analysis of variance (ANOVA) on a mixed effects model with time as a categorical variable as the only fixed effect, patient as the only random effect, and the response as the rank of the score within the patient. Where significance was found with the omnibus test, it was followed with paired Wilcoxon tests comparing each time-point to baseline. Bonferroni adjustment rendered a significance cutoff of .0083 to maintain a 5% false-positive rate.

3.1 Animals

Between November 2013 and August 2017, 17 spayed female dogs were identified for the study (Table S1). At the time of enrollment, the median age was 4.25 years (range, 1.16-10.25). There were 6 Doberman pinscher, 2 mixed breed dogs, and 1 dog of each of the following breeds: standard poodle, giant schnauzer, collie, border collie, American Staffordshire terrier, soft-coated wheaten terrier, Australian shepherd, Labrador retriever, and Puli.

Two dogs were excluded from final data presentation and analysis. In the first dog, an extreme weather event disrupted normal procedures and it was determined by visual inspection of remaining cells immediately after injection that most cells were not viable at the time of injection. The second dog died before the 3-month re-evaluation from acute spinal cord injury, unrelated to the study.

For 15 dogs, the owners reported a median and approximate duration of urinary incontinence of 3 years before enrollment (range, 0.5-9 years); urinary incontinence began at an undetermined time before the remaining 2 dogs being adopted as adults. Medications used that failed to control incontinence before enrollment in the study were PPA (15 dogs), either alone or in combination with diethylstilbestrol (6) or estriol (3). In addition to persistent incontinence, 3 dogs had unacceptable adverse effects that required discontinuation of the medication (aggression with diethylstilbestrol, 1; excessive anxiousness with PPA, 1; vomiting with PPA, 1).

3.2 Isolation and injection of skMPCs

After biopsy, skMPC were expanded in culture for a median of 18 days (range, 14-27 days). For 14 dogs (#4-17), cell passage at the time of injection was tracked: 8 of 14 were injected at passage 2 and 6 of 14 were injected at passage 3. Cells were harvested on the day of injection, and 100 million cells were resuspended in 0.6 mL phenol-free DMEM in preparation for injection. One dog (#13) was injected with only 40 million cells because of slow cell growth.

The surgical procedure was uneventful in all dogs; there were no apparent complications. During the first few injections, it was discovered that the needle could become occluded if the injections were not done rapidly and that successful injections subjectively required a high degree of force. This finding deterred us from transferring this procedure to a minimally invasive technique.

3.3 Assessment of continence

Continence scores increased at some point at the study in 14 of 15 (93.3%) dogs (Figures 1 and 2; Figure S2). The omnibus test was significant (P < .0001). Continence scores were found to be significantly different from baseline at 3 months (P = .006), 6 months (P = .004), 12 months (P = .005), and 24 months (P = .008). Of the 14 dogs with increased continence scores, 7 (50%) achieved a score of 5, 6 (42.8%) achieved a score of 4, and 1 only improved to a score of 3. At the 3-month evaluation, 10 of 15 (66.7%) dogs had improved continence, 4 (26.7%) dogs were unchanged, and 1 dog had worsened incontinence compared to baseline. At the 6- and 12-month evaluations, 13 of 15 (86.7%) dogs had improved continence, 1 dog was unchanged, and 1 dog had worsened incontinence compared to baseline. By 24 months post-injection, 2 dogs had died from unrelated causes and 2 were lost to follow-up. Of the remaining 11 dogs, 9 (81.8%) had improved continence, and 2 were unchanged. Throughout the study, 1 dog never experienced improvement and 3 dogs had initial improvement and then returned to their baseline continence scores. Of the dogs that returned to baseline continence, 2 did so at 12 months and 1 did so at 24 months.

Additional medical management was needed to achieve continence in all but 2 dogs. One of these dogs achieved a continence score of 5 and the other achieved a continence score of 4. Of the remaining 13 dogs, 4 were treated with PPA only, 2 were treated with diethylstilbesterol only, 6 were given a combination of phenylpropanolamine and diethylstilbesterol, and 1 was given a combination of phenylpropanolamine and estriol to improve their continence during the observational period that followed the 3-month post-injection time point. Of the 13 dogs that needed additional medical management, 9 were given the same medications at the end of the study period that were found to be ineffective in controlling their incontinence before skMPC injection; 5 of these 9 had improvement of their incontinence. Two dogs failed to be continent with a combination of phenylpropanolamine and diethylstilbesterol before skMPC injection but were continent with diethylstilbesterol alone after skMPC injection. The remaining 2 of the 13 dogs requiring medication originally were given phenylpropanolamine but by the end of the study were receiving both phenylpropanolamine and diethylstilbesterol to achieve continence.

For 3 dogs, a malfunction in the UPP procedure occurred at baseline, but measurements were continued at other time points; these dogs were excluded from baseline analyses (Figure S3). One dog was not returned for UPP after 3 months and the data were excluded from analysis. The owners of another dog failed to return the dog for UPP at 3 months, but other time points were available for analysis. For a third dog, the owners did not return for the 3- and 6-month time points but other time points were available for analysis. No difference in UPP scores through time was found using an omnibus ANOVA test (P = .41). No significant difference was found at 3 months (P = .12), 6 months (P = .22), 12 months (P = .38), or 24 months (P = .58). A relationship was found between continence score and MUCP (P = .03) with patients achieving higher MUCP values when achieving higher continence scores. In addition, the highest MUCP measured corresponded to a time when the owner reported the highest continence scores in 10 of 14 dogs and a MUCP higher than baseline corresponded to a time when the owner reported higher continence scores in 11 of 14 dogs.