Need to have actual outcome measures finalized to determine n, then d.w. FJR re: stats)
Data analysis: How can you compare before/after if the patients have just had surgery on the affected structures?
Ask Frances for help in coming up with a statistical power test to justify patient numbers.
3.6.1. Sample Size Considerations
This study has a repeated measures structure in which subjects serve as their own controls. Repeated measures will be made prior to implant to confirm that the condition under study is stable or worsening on conventional treatment. Outcome measures obtained at various points during treatment are compared with previous values to determine changes and trends attributable to the treatment or withdrawal of the treatment. A within-arm design will be used to compare prestimulation versus poststimulation changes using paired t-tests or Wilcoxon statistical methods (Table 2 below). For evaluations conducted over the longer term, trends will be studied with conventional regression methods and, where appropriate, by ANOVA according to recommendations of the Statistical Consultation and Research Center, Los Angeles, California.
We have used two outcome measures to perform a power analysis. In each case, the change in the outcome variable due to treatment has been estimated from values in the literature and our own experiences to date with study subjects who are being treated for acute shoulder subluxation. A detailed rationale for the selection of values follows, and the Table below summarizes our predictions quantitatively.
Table 2: Summary of Statistical Plan
Trial Analytic method Primary Site Outcome measure Effect Size SD Source N calc. N+extra
Chronic stroke, shoulder subluxationWe are not using US paired t or Wilcoxon, 1-arm, repeated measures USC Humeral displacement 7mm 7 Faghri, 1994 8 15
The size of each study has been determined using statistical power calculations based upon means and standard deviations of key outcome measures that have been published in previous studies of similar clinical problems and therapies. In these estimations, we assumed that the number of subjects in each arm would be equal and used t-tests to compare continuous variables according to the relationship:
where N = total number of subjects required
Za = the standard normal deviate for a (Za = 1.96 if a = 0.05 and two-tailed and Za = 1.645 if a = 0.05 and one-tailed)
Zb = the standard normal deviate for b (Zb = 0.84 when b = 0.20)
S = standard deviation of the outcome variable, chosen to be conservatively large if more than one value was found in the literature
E = expected effect size
Estimates based on predicate studies, often with different designs and patient profiles, must be interpreted with caution. An increase of sample size by approximately 50% (N+extra) compensates for attrition of subjects, which is common in populations of elderly, sick or community-living individuals. Where no standard deviation of effect size was reported, but standard deviations for stimulated and control subjects were described we have adopted a conservatively large value for standard deviation because we anticipate that chronic subjects may respond more variably to treatment.
The two measures considered to reflect efficacy of treatment best are the amount of joint subluxation before and after treatment Data from Faghri et al. (1994), in a study with a similar design but using surface stimulation in acute patients, suggest a mean improvement in humeral-head displacement of approximately 7mm. After the 6 week treatment period, mean V value of the control group was 9.8 mm ± 6.8 mm whereas the experimental group mean was 2.5 mm ± 3.6 mm. With these values, at least 8 subjects in each group are needed to achieve statistical significance.
We could not find data on muscle atrophy and subsequent remediation by electrical stimulation in stroke survivors. The numbers in table 3 are extrapolated from the results of electrical stimulation of rat muscles using similar parameters. From our preliminary studies in acute stroke survivors, we have found that muscle thickness shows more modest changes than we obtained in rats. There are several reasons why we cannot anticipate the degree of change accurately prior to conducting the experiment. We do not know if the BIONs will recruit all of the muscle or if all study participants will tolerate stimulation that recruits most of the muscle. Our preliminary results in human subjects already suggest that 5% may be too generous an expectation for improvement; some subjects showed this degree of improvement but others showed little change in muscle thickness. Thus, it is likely that more subjects would be required to identify an improvement in shoulder subluxation. We have estimated that at least 15 study participants should be recruited for each of the 2 study groups.
Most of the data analysis will be a comparison between before and after treatment values. Therefore, student t-tests are adequate.
Risks & Benefits
PROCEDURE / Damage to the inferior or superior gluteal neurovascular pedicles
As these patients are already undergoing gluteal rotation flap surgery, both the superior and inferior gluteal neurovascular pedicles will be exposed and easily in reach130. Possible further risks can occur at 2 stages during the procedure:
Once the inferior gluteal pedicle is palpated (the pulsatile artery can be felt easily), the BION will be laid alongside this pedicle under vision, and so some limited elevation of the flap will be required to visualize this (with possibly an additional light source (e.g. a headlight). At no time will the pedicle be opened or further dissected out, other than with the usual normal blunt dissection approach. The BION will be placed with its Tantalum electrode (anode) closest to the inferior gluteal pedicle, and the Iridium electrode can lie either proximally alongside the pedicle or away from the pedicle. Its final position will be tested intra-operatively using an external coil placed outside of a sterile polyurethane drape (so as not to compromise the sterile field at any time), and operated by an assistant.
To prevent the BION from moving in the early post-operative period until the tissues stick back down and the spaces developed close up, it will be affixed via a single 5/0 Vicryl suture around the Tantalum stem at either end of the implant. It will secure the BION to adjacent connective tissue or gluteus maximus fascia, and must be placed so as not to injure the pedicle.
We anticipate that both of these 2 stages together will require minimal increased duration (an increased surgical time of 5-10 minutes at most, in a 1-2 hour procedure); or risk (for patients who are undergoing this surgery already, by a skilled and highly dexterous surgeon, having a wealth of experience with all surgical aspects of this procedure and the relevant anatomy).
The risks during the implantation of the BIONs are the same as for the flap surgery; however, because these patients are already undergoing gluteal rotation flap surgery, the only concern here is the ADDED risk of having the BION implanted during the surgery.
During the gluteal rotation flap surgery, both the superior and inferior gluteal neurovascular pedicles will be exposed and easily in reach130. Possible further risks can occur at 2 stages during the procedure:
Once the inferior gluteal pedicle is palpated (the pulsatile artery can be felt easily), the BION will be laid alongside this pedicle under vision, and so some limited elevation of the flap will be required to visualize this (with possibly an additional light source (e.g. a headlight). At no time will the pedicle be opened or further dissected out, other than with the usual normal blunt dissection approach. It is unlikely, but possible, that the artery and or nerve will be mistakenly cut or damaged.
To prevent the BION from moving in the early post-operative period until the tissues stick back down and the spaces developed close up, it will be affixed via a single 5/0 Vicryl suture around the tantalum stem at either end of the implant. It will secure the BION to adjacent connective tissue or gluteus maximus fascia, and must be placed so as not to injure the pedicle. It is unlikely, but possible, that tissue will be damaged during the suturing.
We anticipate that both of these 2 stages together will require minimal increased duration (an increased surgical time of 5-10 minutes at most, in a 1-2 hour procedure); or risk (for patients who are undergoing this surgery already).
There is a small possibility that a foreign body reaction to the device may occur; in this case, the device could be removed in a short procedure.
A possible benefit for subjects in this study is the possible decrease in pressure wound recurrence, as well as better gluteal muscle health (including increased muscle size and increased blood perfusion) on the stimulated side.
A possible benefit for society is a better understanding of stimulation parameters for this particular application, as well a better understanding of wound healing in general.
Insert Budget Excel Sheet here
100. Levine SP, “Functional Electrical Stimulation for Pressure Sore Inhibition”, USPTO Patent 4,727,878, 1988 March 1 (filed 1985 Sept 26; Assignee: The University of Michigan, Ann Arbor, Mich.).
101. National Decubitus Foundation, “Cost Savings Through Bedsore Avoidance”, National Decubitus Foundation, 4255 S. Buckley Rd., Suite 228, Aurora, CO 80013 (www.decubitus.org/cost/cost.html).
102. Wells C & Hooker S, “The Spinal Injured Athlete”, Adapted Physical Activity Quarterly, 7:265-285, 1990.
103. Rimmer J, “Fitness and Rehabilitation Programs for Special Populations”, p.238, Brown and Benchmark, Madison, WI, 1994.
104. Bogie KM et al, “Electrical Stimulation for Pressure Sore Prevention and Wound Healing”, Asst Technol 12:50-66, 2000.
105. Bogie KM et al, “Improving the Health of Paralyzed Tissue using Electrical Stimulation”, U.S. Department of Veterans Affairs, Rehabilitation Research & Development Service, 3rd National Rehabilitation R&D Meeting, www.vard.org/va/02/htm/rrds_feb_2002_confbogie1.htm, Feb 2002.
107. HK DCES IEEE Paper.
108. Fourth National PU Prevalence Survey.
110. HK dissection.
111. Fuhrer, Garber et al '93.
112. National Pressure Ulcer Advisory Panel, “Pressure Ulcers: Incidence, economics, Risk Assessment”, Consensus Development Conference Statement, West Dundee, Illinois, S-N Publications Inc., 1989.
113. Stacy RW, "Health & Economic Benefits of FES Induced Active Physical Therapy", 1986.
114. 1 in every 3.6 immobile person develop pressure ulcers (Nick).
124 National Pressure Ulcer Advisory Panel, “Clinical Practice Guideline on Pressure Ulcers in Adults: Prediction and Prevention”, www.npuap.org.
125. Wilhelmi BJ & Neumeister M, "Pressure Ulcers, Surgical Treatment and Principles", eMedicine, www.emedicine.com/plastic/topic462.htm, Feb 14 2002.
126. U.S. Department of Health and Human Services, "PU Treatment - Quick Reference Guide for Clinicians", AHCPR Publication No. 95-0653, Rockville, MD, Dec 1994.
127. Dharmarajan TS & Ahmed S, “The Growing Problem of Pressure Ulcers – Evaluation & Management for an Aging Population”, Postgraduate Medicine, Vol. 113,5, May 2003.
128. Bergstrom N et al, "Treatment of Pressure Ulcers. Clinical Practice Guideline No. 15", US Dept of Health and Human Services, Agency for Health Care Policy and Research, Rockville, MD, AHCPR Publication 95-0652, www1.neweb.ne.jp/wb/decubitus/Clinical_Practice_Guideline.htm, 1994.
129. Yarkony GM, “Pressure Ulcers: A Review”, Arch Phys Med Rehabil, 75:908-917, Aug 1994.
130. McGraw JB & Arnold PG, “McGraw & Arnold’s Atlas of Muscle and Musculocutaneous Flaps”, Hampton Press Publishing Company, Inc., Norfolk, VA, 1986.
131. Mathes SJ & Nahai F, “Clinical Atlas of Muscle and Musculocutaneous Flaps”, CV Mosby Company, St Louis, MO, 1979.
132. NGC Guideline, "Pressure Ulcer Prevention and Treatment Following Spinal Cord Injury", National Guideline Clearinghouse (evidence-based clinical practice guidelines), Agency for Healthcare Research & Quality, U.S. Department of Health & Human Services, American Medical Association, American Association of Health Plans, www.guideline.gov/summary/summary.aspx?doc_id=2589&nbr=1815, Dec 2003.
133. Disa JJ, Carlton JM, Goldberg NH, “Efficacy of Operative Cure in Pressure Sore Patients”, Plastic & Reconstructive Surgery, 89(2):272-8, Feb 1992; Comment in: Plast Reconstr Surg. 1992 Nov;90(5):930.
134. Olshansky K, “Essay on Knowledge, Caring, & Psychological Factors in Prevention & Treatment of Pressure Ulcers”, adapted from Panel Discussion at 2nd Annual Wound Healing Symposium, Medical College of Virginia, Richmond, VA, www.woundheal.com/pubs/bedsorePub01.htm, May 1993.
135. Vohra RK & McCollum CN, “Fortnightly Review: Pressure Sores”, British Medical Journal, 309:853-857, http://bmj.bmjjournals.com/cgi/content/full/309/6958/853#R64, Oct 1994.
136. Relander M, Palmer B. Recurrence of surgically treated pressure sores. Scand J Plast Reconstr Surg 1988;22:89-92.
137. Luscher NJ, de Roche R, Krupp S, Kuhn W, Zach GA. The sensory tensor fasciae latae flap: a 9-year follow-up. Ann Plast Surg 1991;26: 306-11.
138. Leasavoy MA, Dubrow TJ, Korn HN, Cedars MG, Castro DJ. "Sensible" flap coverage of pressure sores in patients with meningomyelocele. Plast Reconstr Surg 1990;85:390-4.
150. National Pressure Ulcer Advisory Panel, "Pressure Ulcers in America: Prevalence, Incidence and Implications for the Future", Cuddigan, J., Ayello, E.A., & Sussman, C. (Eds.), Reston VA: NPUAP, 2001.