Use of Localized Human Growth Hormone/Testosterone Injections for Chronic Shoulder Pain: A Case Series, with 12-Month Follow-up

Background

Chronic shoulder pain is a common condition which involves long term pain and disability. Common problems associated with chronic shoulder pain involve tears and fraying of the rotator cuff tendons, the labrum, and involvement of the joint capsule. Most of these conditions are due to degeneration or injury to dense connective tissue, primarily composed of ground substance and collagen. Recent literature indicates that “micro instability” may play a role in the development of glenohumeral pathology including tendinopathy and labral lesions [1-3]. An intervention that would lead to intrinsic repair and strengthening of the involved dense connective tissues would be expected to have benefit in regard to pain and an improvement in function.

The Orth biologic/Regenerative Medicine revolution is expanding treatment for chronic pain. Chronic musculoskeletal conditions involving joint, tendon, ligament and cartilage are the primary targets. Regenerative medicine involves the use of bioactive cells or factors to stimulate healing and remodeling of pathological tissues. The most commonly studied procedures have been proliferative therapy [4-9], injections with platlet-rich plasma. [4,6,7,10, 11], and increasingly, stem cell injections [12] and other tissue engineering interventions [13].

An emerging area of regenerative medicine is the use of human growth hormones and other anabolic agents. A few numbers of both animal and human studies have been published, with somewhat mixed results regarding the efficacy of these hormones in facilitating injury repair on musculoskeletal tissues [6,11,14-18]. Demonstrating the ability for the use of potent anabolic agents to improve clinical outcomes in the management of diagnosed chronic musculoskeletal conditions can be a significant advancement in the treatment of these conditions.

In a previously reported study involving the treatment of chronic low back pain, localized injections of recombinant human growth hormone (rhGH) and testosterone were found to lead to an improvement in pain and function [19]. It was hypothesized that this was due at least in part to an improvement in the integrity and strength of the treated ligaments, tendons, and joint capsules. RhGH and testosterone are both endogenous anabolic hormones that stimulate protein synthesis [20,21] Research describing the mechanism of action of rhGH and testosterone has been previously described [19] It is felt that the localized injection of rhGH and testosterone interact synergistically at the injection site to stimulate protein synthesis, leading to formation of new collagen and ultimately mature connective tissue. In conjunction with the injections, controlled exercise would be expected to positively influence the remodeling of the dense connective tissue, while possibly preventing contractures.

Case presentations

This is a 12-month follow-up case series using a one-group pretest posttest design regarding patients seen in a private pain management clinic. This study is designed to provide preliminary data and determine if a randomized control trail is warranted in the future.

Participants received a standardized shoulder examination by an anesthesiologist/pain management physician. Participants had chronic shoulder pain, with the examination suggesting primarily capsular/tendon or labral involvement with or without associated acromioclavicular joint or rotator cuff involvement in the production of symptoms.

The study was approved by and performed under the auspices of the Institutional Review Board (IRB) of Roper St. Francis Healthcare, Charleston, SC, USA. Data were monitored by a Data Collection Site Monitoring Board and Adverse Experience Reporting Committee. All participants provided informed consent. The outcomes and intervention data used to support the findings of this study are restricted by HIPAA in order to protect patient privacy. Data are available from the primary author for researchers who meet the criteria for access to confidential data and with approval from the IRB of Roper St. Francis Healthcare. No external funding was involved with this case series.

Participants

Table 1: Inclusion and exclusion criteria.

Participants were 23 consecutive eligible adult participants, aged 18-85 with chronic shoulder pain, serving as a convenience sample during the interval of June 2009 through June 2011. Inclusion and exclusion criteria are listed in Table 1.

Five participants had undergone previous shoulder surgery, with one having undergone two previous surgeries. The surgeries included three for rotator cuff repair, two for arthroscopy and debridement including a SLAP lesion repair. Seven of the shoulders had a history of previous dislocations, and two participants had been diagnosed with hypermobility syndrome. The participants had received prior therapies including trigger point injections, subacromial and gleno-humeral joint steroid injections, physical therapy, chiropractic adjustments, and pharmacological pain management. The previoius interventions were not consistent across participants. Participant’s demographics are in Table 2.

Table 2: Demographics for participants completing study.

The examination included active/passive range of motion in abduction and flexion, grip strength, Obrien’s Active Compression test, glenohumeral external rotation apprehension test, rotator cuff testing for weakness and/or pain, deltoid testing for weakness and/ or pain, passive horizontal adduction of the glenohumeral joint for acromioclavicular joint (ACJ) pain, palpation of the coracoid process, ACJ, bicipital groove, greater tuberosity, and bicep and triceps strength testing.

To be included in the cohort, the Obrien’s Active Compression test and/or the apprehension test demonstrated reproduction of familiar symptoms or aberrant motion. Local anesthetic injections were performed under fluoroscopic guidance in the region of tenderness and suspected pathology. When the ACJ or rotator cuff findings were positive, injections were performed at the ACJ capsule, coraco-clavicular ligament, or the region of the humeral attachment of the affected rotator cuff muscle. Subjective relief of symptoms by 50% or more and improved strength testing indicated continuation in the study. All patients included met this criterion./p>

Intervention

Each participant meeting the inclusion criteria received a series of injections of rhGH/testosterone to the same areas as the local anesthetic was injected. The injections were performed by a board-certified, hospital-affiliated anesthesiologist/pain management physician under fluoroscopic guidance.

The injections consisted of recombinant rhGH (Sandoz International, Holzkirchen, Germany, 3u/10 cc volume), aqueous testosterone (12.5 mg for females, 25 mg for males, per 10 cc volume), 5 cc of 1% preservative-free procaine, and 0.9% NaCl to a total volume of 10 cc. The injections were performed at 3–4-week intervals and averaged 3-4 in number. Approximately 3, 10 cc syringes were administered at each injection visit. The number of injections needed for each participant was determined by improvements noted in Obrien’s and apprehension tests, as well as the tissue feel during injections. It is noted by the primary author that healthy and healing ligament and tendon tissue presented with a dense feel to syringe pressure, while degenerative/weakened ligament or tendon tissue has more of a soft feel.

Once able, the participants performed graded resisted exercise as tolerated and range of motion activities. Individualized exercise programs included the following:

a) Pilates exercise with a certified instructor

b) Physical therapy for range of motion, motor control and resistance exercise

c) Self-administered treatment at a fitness center or home

d) Working with a personal trainer at an exercise facility

e) After completion of the injections, the participants were followed for 12-months.

Outcome Measures

The primary outcome was determined by the Shoulder Pain and Disability index (SPADI) [22]. Secondary outcomes were the Shoulder Pain Score (SPS), [23] the patients pain rating from 0-10 using the Mankowski Pain Scale (MPSI), [24] and the participants reported overall percentage change in presenting symptoms [25]. The SPADI has been used to measure self-rated function across a number of shoulder diagnoses [26-30] The questionnaire is based on a score of 100, with the minimal detectible change (MDC 95%) described as 18 points [31].

The Shoulder Pain Score assesses the patient’s self-rated shoulder pain. This pain score has been found to provide a reliable rating of pain and is useful to follow the improvement in pain from the initial visit to the post treatment interval [32].

The MPSI was used as the pain rating scale. It is a 0-10 pain rating scale that links written descriptors of the intensity of the pain to a numerical value. It has been shown to be a valid and reliable tool for pain measurement, as compared to the Faces Scale, the Visual Analog Scale, and the Numeric Rating Scale [24]. The minimal clinically important difference for the change score has not been determined, but it has been reported for the Numeric Rating Scale as two points [33].

Testosterone levels were measured as total testosterone, and rhGH levels measured as IGF-1 at the beginning of the study in order to determine if baseline blood levels were a positive variable related to the outcome measures.

Statistical Analysis

All analyses were performed using SAS version 9.4 (Cary, NC). Statistical significance was defined at α=0.05. Descriptive statistics were used to compare the characteristics of the study sample. Means and standard deviations were calculated to describe continuous data. Frequencies and percentages were calculated to describe categorical data. An intention to treat analysis was used, with the last value carried forward for the patients who did not complete the injections.

The primary outcome measure was a change in SPADI scores. Secondary outcome measures were changes in SPS and MPSI scores. Changes in scores from baseline were assessed for normality visually using QQ-plots and statistically using the Shapiro-Wilk test. All three instrument scores were found to be normally distributed: SPADI (p=0.3531), SPS (p=0.0880), and MPSI (p=0.1243). Changes in scores were tested using paired Student’s t-tests.

Results

Raw outcome data is reported in Tables 3, 4, and 5. Of the 23 eligible patients, six chose not to enter the treatment program. Figure 1. Of the remaining 17 eligible participants, all provided informed consent to enter the study. All phases of the study including the 12-month follow-up were completed by 14 participants (16 shoulders). Of the three who did not complete the study, one completed injection but started another injection protocol near the six-month follow-up. The patient did not feel that progression was quick enough. One underwent surgery prior to completing the injections, and one underwent surgery shortly after completing the injections. Both surgical procedures were performed because of continued pain and decrease function, related to the original complaint.

Table 3: Intake and post-treatment data.

Note*: Duration is the duration of shoulder pain for the present episode. The percent self-rated improvement is the patient’s perception of improvement at 12-months following the first injection. BMI is body mass index.

Table 4: Standardized Outcomes.

Note*: SPS is the Shoulder Pain Score. SPADI is the Shoulder Pain and Disability Index score. MPS is the Mankowski Pain Scale score. Side is the side of the body receiving injections.

Table 5: Human growth hormone, insulin-like growth factor and testosterone levels.

Note*: Mean HGh/inj and Test/inj volumes of HGh and testosterone used with injections. The patient reported events related to the development of shoulder pain. IGF-1 is the baseline level of insulin like growth factor with blood testing, and Testos is the base line level of testosterone. No correlation was seen with base line blood levels and results.

Figure 1: Consort diagram for study procedures.

Across the19 shoulders (17 patients) at the 12-month follow- up, statistically significant mean decreases in Shoulder Pain and Disability Index, Shoulder Pain Score and pain severity (Mankowski Pain Scale) were found. There was also a significant improvement in the participants’ report of symptoms with an overall self-improvement of 76%.

The primary outcome measure of change in SPADI was found to significantly change from baseline, with a mean reduction in score of 31.8 points (95% Confidence Interval (CI): 20.7-43.0; p<.0001). Figure 2. The SPS also changed from baseline, with a mean reduction in score of 36.7 points (95% CI: 24.0-49.4, p<.0001). Figure 3. Finally, the MPSI showed a change from baseline, with a mean change in score of 3.1 points (95% CI: 2.0-4.2, p<.0001). Figure 4.

Note*: Figure 2 Shown above are the point estimates for the mean scores and 95% confidence limits for the SPADI scores, which shows a mean reduction of 31.8 points (95% CI: 20.7-43.0, p<.0001). The minimal detectable change is 18.1 points, with a minimal clinically important difference of 13.2 points (Schmitt & Di Fabio, 2004).

Figure 2: Shoulder Pain and Disability Index (SPADI) scores pre- and post-test.

Note*: Figure 3. The point estimates for the mean scores and 95% confidence limits for the SPS scores, which shows a mean reduction of 36.7 points (95% CI: 24.0-49.4, p<.0001).

Figure 3: Shoulder Pain Scores (SPS) scores pre- and post-test.

Note*: Figure 4. The point estimates for the mean scores and 95% confidence limits for the MPSI scores, which shows a mean reduction of 3.1 points (95% CI: 2.0-4.2, p<.0001).

Figure 4: Mankowski Pain Scale Index (MPSI) scores pre- and post-test.

There were no significant correlations between study entry IGF-1 or testosterone levels and changes in the four outcome variables at 12-months. No adverse reactions related to the study were reported during the course of the treatment or at the follow-up sessions.