Copyright © 2009 by The Journal of Bone and Joint Surgery, Inc.

Commentary & Perspective

Commentary & Perspective on
"Outcome of Lumbar Arthrodesis in Patients Sixty-five Years of Age or Older"
by Steven D. Glassman, MD, et al.

Commentary & Perspective
"Deconstructing Pieces: Risks of Subgroup Reporting for Regulatory Clinical Trials"
by Sohail K. Mirza, MD, MPH*,
Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire

Posted April 2009

All the king's men could not restore the pieces to make Humpty Dumpty whole again. The same fate can befall subgroup analyses of clinical trials, if these analyses fail to respect the boundaries set by the original full study. Successfully completed clinical trials are the product of great effort from clinical investigators, substantial investment by study sponsors, and remarkable generosity of participating subjects. They are repositories of valuable data. Every effort should be made to utilize these data maximally, and subgroup analyses of these data can yield useful information. However, when interpreting the results for subsets of the original study population, the context and the results of the full study must always be considered.

In their paper, "Outcome of Lumbar Arthrodesis in Patients Sixty-five Years of Age or Older," Glassman et al. present an analysis of data for the control arm of a clinical trial being performed for the purpose of seeking regulatory approval for a spinal fusion device. The analysis was performed in response to a call for evidence by the Centers for Medicare and Medicaid Services (CMS). Medicare is the largest payer of health services in the United States, and analyses for coverage decisions by its powerful Medicare Coverage and Advisory Committee (MCAC) get attention, both from other payers who follow Medicare's lead and also from those who receive payment for health services. In November 2006, The MCAC identified a lack of evidence for safety and efficacy for lumbar fusion in older adults. Glassman et al. have addressed this challenge by comparing results for older and younger adults in the Food and Drug Administration Investigational Device Exemption (FDA-IDE) trial for use of recombinant human bone morphogenetic protein-2 (rhBMP-2) in posterolateral lumbar arthrodesis.

In the subgroup analysis presented by Glassman et al., 224 subjects in the control arm of an FDA-IDE study were divided into two groups on the basis of age (subjects who were less than sixty-five years of age, and subjects who were sixty-five years of age or older). Pain and function outcomes were noted to be equivalent or better in older adults, despite a higher frequency of adverse events. The study seems to provide compelling evidence in support of lumbar fusion. It has the requisites for an excellent clinical study: It is a prospective study with predefined end points; the trial was registered with the national clinical trials database1; the characteristics of the study population were broadly measured at baseline; surgery was uniformly administered; outcomes were measured with use of validated instruments; a high follow-up rate was achieved; and safety was measured meticulously. In addition, since this was a regulatory trial, external surveillance for adherence to the study protocol was ensured by the FDA, and the analyses performed were sophisticated and thorough. A section in the manuscript also describes study sponsorship, and the authors additionally disclose other support provided to them by the device manufacturer. We should have high confidence in the findings.

However, two problems related to potential selection bias somewhat limit interpretation of this study. First, the clinical indications for the surgery are unclear. The participant inclusion and exclusion criteria in the study protocol registered with ClinicalTrials.gov1 (Table I of this Commentary) describe enrolling patients with back pain associated with single-level lumbar disc degeneration: a controversial diagnosis commonly referred to as "discogenic back pain." However, the authors report in their Table II that 88% to 90% of subjects had a diagnosis of disc herniation, and 10% to 14% had instability. Results for surgical treatment of disc herniation and spondylolisthesis are generally more favorable than are results for surgical treatment of discogenic back pain2. Another exclusion criterion listed in this paper, risk of noncompliance with study protocol, is also potentially a problem because patients lost to follow-up may be excluded on the basis of this criterion; it is not listed in the more detailed list of participant exclusions in the registration with ClinicalTrials.gov. Similar to the outcomes reported in other U.S. FDA trials for surgical treatment of discogenic back pain3,4, the outcomes reported in this study are better than outcomes reported in three European randomized trials5-7 (Table II of this Commentary). The better results may, in part, be due to selective inclusion of patients with structural causes for back and leg pain and selective exclusion of noncompliant patients.

Table I
Participant Inclusion and Exclusion Criteria Specified in the Study Protocol Registered with Clinicaltrials.gov Service of the National Institutes of Health1

Inclusion criteria: each patient participating in this clinical trial must meet all of the following inclusion criteria:
 

1. Has degenerative disc disease as noted by back pain of discogenic origin, with or without leg pain, with degeneration of the disc confirmed by patient history of pain and radiographic studies

2. Requires fusion of a single-level disc space from L1 to S1

3. Has not responded to nonoperative treatment (e.g., bed rest, physical therapy, medications, spinal injections, manipulation, and/or transcutaneous electrical nerve stimulation [TENS]) for a period of 6 months

4. If of child-bearing potential, patient is nonpregnant, non-nursing, and agrees to use adequate contraception for 1 year following surgery

Exclusion criteria: a patient meeting any of the following criteria is to be excluded from this clinical trial:
 

1. Has primary diagnosis of a spinal disorder other than degenerative disc disease with Grade 1 or less spondylolisthesis at the involved level

2. Had previous spinal fusion surgical procedure at the involved level

3. Requires spinal fusion at more than one lumbar level

4. Has been previously diagnosed with osteopenia

5. Has presence of active malignancy or prior history of malignancy (except for basal cell carcinoma of the skin)

6. Has a history of autoimmune disease (e.g., systemic lupus erythematosus or dermatomyositis)

7. Has a history of exposure to injectable collagen or silicone implants

8. Has received treatment with an investigational therapy within 28 days prior to implantation surgery or such treatment is planned during the 16 weeks following rhBMP-2/CRM implantation

9. Has received any previous exposure to any or all BMPs of either human or animal extraction

Table II
Change in Oswestry Disability Index Scores Reported in Randomized Trials of Single-Level Lumbar Fusion for Treatment of Back Pain Associated with Degenerative Disc Disease
 

European Trials

United States FDA Trials

Fritzell et al.6

Brox et al.7

Fairbank et al.5

BAK-fusion3

360-fusion4

Glassman et al.

<65 years old

≥65 years old

Number of patients in surgical group

201

35

176

99

80

174

50

Mean baseline Oswestry score (points)

47

42

46

52

63

Not reported

Not reported

Mean final Oswestry score

35

26

34

31

40

Not reported

Not reported

Mean final difference and % reduction

12 (24%)

16 (37%)

12 (26%)

21 (41%)

23 (37%)

24.5

28.5

Second, only selected safety outcomes are reported. Although pain and function results were generally similar for the two age groups, adverse events were distinctly more frequent in the older age group. The safety data yielded by this study may be particularly useful because product safety assessment is the primary focus in FDA non-inferiority or equivalence trials. The FDA performs detailed safety monitoring in IDE studies. Despite the importance assigned to safety assessment, four available sources of safety data for this study are not easy to reconcile. In this paper, adverse events were identified in 56% of subjects who were sixty-five years of age or older and in 36% of subjects who were less than sixty-five years of age. Rates for all adverse events were not reported in the original publication of the FDA-IDE results8, and higher total adverse event rates (86% to 92%) are reported in the safety summaries available in the FDA web site for both the pilot study and the overall study9 (Table III, a, b, and c of this Commentary). The reported differences are likely related to differences in selection and definitions of adverse events. As is often the case, the investigators have selected those adverse events that they believe are important; but important adverse events are those that matter to patients, and until we know more about what matters to patients, the safer approach may be to report all adverse events tracked in the FDA data. Knowledge of the rates of specific adverse events may be crucial for informed choice in elective back surgery. Some patients may decline surgery purely on account of the risk of adverse events, regardless of the potential for reduction of pain.

Table III
Number of Subjects and Selected Safety Outcomes Reported to the FDA for Parent rhBMP-2 Posterolateral Fusion Investigational Device Exemption and Humanitarian Device Exemption Studies.
 

(a) Published Report of FDA-IDE Study8

Investigational Arm

Control Arm

Number of subjects

53

45

Total number of adverse events

61

48

Any adverse event

Not reported

Not reported

Death

Not reported

Not reported

Reoperation

Not reported

Not reported

Dural injury

3 (5.7%)

5 (11.1%)

Infection

0

1 (2.2%)

Neurological

Not reported

Not reported

Nonunion

0

2 (9.5%)
 

(b) FDA Summary for Posterior Fusion Pilot Clinical Trial (INFUSE Bone Graft/MASTERCRAFT Granules/CD Horizon Spinal System)9

Investigational Arm

Control Arm

Number of subjects

25

21

Total number of adverse events

74

69

Any adverse event

22 (88%)

18 (85.7%)

Death

0

0

Reoperation

3 (12.0%)

2 (9.5%)

Dural injury

2 (8.0%)

2 (9.5%)

Infection

4 (16.0%)

5 (23.8%)

Neurological

4 (16.0%)

3 (14.3%)

Nonunion, pending

0

2 (9.5%)

Nonunion, failure

1 (4.0%)

2 (9.5%)
 

(c) FDA Summary for Posterior Fusion (rhBMP-2/Compression Resistant Matrix/CD Horizon Spinal System)9

Investigational Arm

Control Arm

Number of subjects

239

224

Total number of adverse events

740

696

Any adverse event

217 (90.8%)

207 (92.4%)

Death

3 (1.3%)

4 (1.8%)

Reoperation

12 (5.0%)

11 (4.9%)

Dural injury

14 (5.8%)

18 (8.0%)

Infection

47 (19.7%)

46 (21.4%)

Neurological

80 (33.5%)

69 (30.8%)

Nonunion, pending

5 (2.1%)

6 (2.7%)

Nonunion, failure

6 (2.5%)

17 (7.5%)

Glassman et al. provide a helpful subgroup analysis of an FDA trial, but the original study population needs to be clearly identified and different versions of slicing data must, in the end, add up to the data contained in the original study. Including the current study by Glassman et al., as many as seven recently published papers have described segments of the original study. From these separate reports, it is difficult to reconstruct the original patient sample, predefined analysis plans, or the outcomes thresholds used in the original sample-size power calculations10-15. Safety results differ among the reports. Safety outcomes are perhaps the most reliable aspects of FDA studies, and since safety concerns are often at the forefront of our discussions with patients, the FDA could greatly help patients make informed treatment choices by simplifying public reports of approved product safety and efficacy data.

*The author did not receive any outside funding or grants in support of his research for or preparation of this work. Neither he nor a member of his immediate family received payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity.

References

1. ClinicalTrials.gov. rhBMP-2/CRM/CD HORIZON® Spinal System Pivotal Study. http://www.clinicaltrial.gov/ct2/results?term=NCT00707265&show_down=Y#down. Accessed 2009 Feb 13.
2. Weinstein JN, Lurie JD, Tosteson TD, Skinner JS, Hanscom B, Tosteson AN, Herkowitz H, Fischgrund J, Cammisa FP, Albert T, Deyo RA. Surgical vs nonoperative treatment for lumbar disk herniation: the Spine Patient Outcomes Research Trial (SPORT) observational cohort. JAMA. 2006;296:2451-9.
3. Blumenthal S, McAfee PC, Guyer RD, Hochschuler SH, Geisler FH, Holt RT, Garcia R Jr, Regan JJ, Ohnmeiss DD. A prospective, randomized, multicenter Food and Drug Administration investigational device exemptions study of lumbar total disc replacement with the CHARITE artificial disc versus lumbar fusion: part I: evaluation of clinical outcomes. Spine. 2005;30:1565-75.
4. Zigler J, Delamarter R, Spivak JM, Linovitz RJ, Danielson GO 3rd, Haider TT, Cammisa F, Zuchermann J, Balderston R, Kitchel S, Foley K, Watkins R, Bradford D, Yue J, Yuan H, Herkowitz H, Geiger D, Bendo J, Peppers T, Sachs B, Girardi F, Kropf M, Goldstein J. Results of the prospective, randomized, multicenter Food and Drug Administration investigational device exemption study of the ProDisc-L total disc replacement versus circumferential fusion for the treatment of 1-level degenerative disc disease. Spine. 2007;32:1155-63.
5. Fairbank J, Frost H, Wilson-MacDonald J, Yu LM, Barker K, Collins R; Spine Stabilisation Trial Group. Randomised controlled trial to compare surgical stabilisation of the lumbar spine with an intensive rehabilitation programme for patients with chronic low back pain: the MRC spine stabilisation trial. BMJ. 2005;330:1233.
6. Fritzell P, Hägg O, Wessberg P, Nordwall A; Swedish Lumbar Spine Study Group. Lumbar fusion versus nonsurgical treatment for chronic low back pain: a multicenter randomized controlled trial from the Swedish Lumbar Spine Study Group. Spine. 2001;26:2521-34.
7. Brox JI, Sørensen R, Friis A, Nygaard Ø, Indahl A, Keller A, Ingebrigtsen T, Eriksen HR, Holm I, Koller AK, Riise R, Reikerås O. Randomized clinical trial of lumbar instrumented fusion and cognitive intervention and exercises in patients with chronic low back pain and disc degeneration. Spine. 2003;28:1913-21.
8. Dimar JR, Glassman SD, Burkus KJ, Carreon LY. Clinical outcomes and fusion success at 2 years of single-level instrumented posterolateral fusions with recombinant human bone morphogenetic protein-2/compression resistant matrix versus iliac crest bone graft. Spine. 2006;31:2534-40.
9. Food and Drug Administration. Summary of Safety and Probable Benefit: recombinant human bone morphogenetic protein-2 (rhBMP-2) contained on an absorbable collagen sponge (ACS) combined with a calcium phosphate bone void filler bulking agent. Notice of final approval, 2008 Oct 10. http://www.fda.gov/cdrh/pdf4/h040004b.pdf. Accessed 2009 Feb 13.
10. Carreon LY, Glassman SD, Djurasovic M, Campbell MJ, Puno RM, Johnson JR, Dimar JR 2nd. RhBMP-2 versus iliac crest bone graft for lumbar spine fusion in patients over 60 years of age: a cost-utility study. Spine. 2009;34:238-43.
11. Glassman SD, Carreon LY, Djurasovic M, Campbell MJ, Puno RM, Johnson JR, Dimar JR. RhBMP-2 versus iliac crest bone graft for lumbar spine fusion: a randomized, controlled trial in patients over sixty years of age. Spine. 2008;33:2843-9.
12. Glassman SD, Carreon LY, Campbell MJ, Johnson JR, Puno RM, Djurasovic M, Dimar JR. The perioperative cost of Infuse bone graft in posterolateral lumbar spine fusion. Spine J. 2008;8:443-8.
13. Carreon LY, Glassman SD, Brock DC, Dimar JR, Puno RM, Campbell MJ. Adverse events in patients re-exposed to bone morphogenetic protein for spine surgery. Spine. 2008;33:391-3.
14. Glassman SD, Dimar JR 3rd, Burkus K, Hardacker JW, Pryor PW, Boden SD, Carreon LY. The efficacy of rhBMP-2 for posterolateral lumbar fusion in smokers. Spine. 2007;32:1693-8.
15. Glassman SD, Carreon LY, Dimar JR, Campbell MJ, Puno RM, Johnson JR. Clinical outcomes in older patients after posterolateral lumbar fusion. Spine J. 2007;7:547-51.