J.K. Burkus, M.F. Gornet, T.C. Schuler, T.J. Kleeman, and T.A. Zdeblick reply:
Our paper presented the long-term InFUSE/LT-CAGE outcomes data collected from 277 patients, including 143 patients managed with rhBMP-2 through an open approach who were enrolled in a randomized controlled trial1 and 134 patients managed with rhBMP-2 through a laparoscopic approach who were enrolled in a concurrent, nonrandomized, single-arm, prospective trial. In total, as reported to the FDA2, 288 patients managed with rhBMP-2 and 139 control patients managed with iliac crest bone graft were enrolled in these two trials, along with an additional eleven patients managed with rhBMP-2 and three patients managed with iliac crest bone graft from a pilot study3. In the FDA labeling cohort of patients, eleven retrograde ejaculation events were listed in the rhBMP-2 group, and one retrograde ejaculation event was listed in the iliac crest bone graft control group. Four important points must be made.
First, there is no relationship between the use of rhBMP-2 in stand-alone interbody fusion cages and the postoperative development of retrograde ejaculation. In the prospective randomized controlled trial comparing the use of rhBMP-2 with autograft, the retrograde ejaculation rates in male patients were 6.4% (five of seventy-eight) and 1.5% (one of sixty-eight), respectively. These rates are not significantly different on the basis of treatment (p = 0.216, Fisher exact test). Previous reports in the literature and experience have established that the laparoscopic (transperitoneal) technique increases the risk of retrograde ejaculation4-9. Therefore, it is not surprising that the retrograde ejaculation rate would be increased when a nonrandomized group of 134 patients managed with the laparoscopic technique was included.
In a more recent randomized controlled trial10, a control group of patients managed with rhBMP-2 with the LT-CAGE device was compared with another group of patients managed with an investigational artificial lumbar disc, with all procedures being performed through an open approach. Retrograde ejaculation was observed in 2.3% (two) of eighty-six male patients who received the rhBMP-2 treatment, compared with 2.0% (four) of 205 male patients who received the lumbar disc treatment. These rates were similar (p = 1.000), which further confirms that there is no relationship between the use of rhBMP-2 treatment and the occurrence of retrograde ejaculation.
Second, there is no relationship between the use of rhBMP-2 and the late development of retrograde ejaculation. No new retrograde ejaculation events were reported during the follow-up assessments between twelve months and seventy-two months. In a 2003 report4, the use of a transperitoneal approach to the lumbosacral spine, not the use of rhBMP-2, was established as a predisposing factor in the development of retrograde ejaculation. This conclusion is also widely supported in the literature5-9. In the subset of patients reported by Sasso et al.4, retrograde ejaculation symptoms resolved in two of six patients. In the total group of 288 patients managed with rhBMP-2, the retrograde ejaculation events resolved in five of the eleven patients and were still ongoing in the other six patients when they were last seen at forty-eight months (two patients) or seventy-two months (four patients) of follow-up.
Third, late retrograde ejaculation is not associated with anterior lumbar bone formation. Local, exuberant bone formation posterior to the disc space may cause neurologic compression. The spinal canal is a closed area bounded by the pedicles, facet joints, lamina, and soft tissues. In contrast, the retroperitoneal space anterior to the disc is not a closed area. Radial osteophytes commonly occur in degenerative lumbar spondylosis adjacent to the disc level. Posterior neurologic compromise resulting from the development of these radial osteophytes commonly results in neurogenic claudication and lumbar radiculopathies, but retrograde ejaculation has not been reported to result from these anterior osteophytes.
Fourth, there is no clinical evidence of marked anterior bone formation in this series. The comment made by Dr. Smoljanovic and colleagues refers to a figure from a 2000 pilot study3, which shows that the cages are placed flush with the anterior cortex of the vertebral bodies. The cages are not recessed within the disc space. There is a thin, shallow rim of bone covering the cages and not extending into the soft tissues of the retroperitoneal space. New bone formation commonly occurs within 1 to 2 mm of cages filled with rhBMP-2. This is not ectopic bone formation.
Radiographic reviewers did not assess anterior bone formation in the studies. Treating physicians identified no symptoms related to anterior bone formation. There were no reoperations or reported complications resulting from anterior bone formation. The senior author (J.K.B.) reviewed the radiographs and computed tomography scans for all eleven patients with complaints of retrograde ejaculation. Two patients had both cages recessed >3 mm inside the disc space, two patients had cages prominently placed anterior to the disc space, and seven patients had cages placed flush with anterior cortical margins of the disc space. Five of the eleven patients had a thin shell of new bone formation covering the cages. None showed any bone extending into the soft tissues.