The Ertl Controversy: Follow-up of Patients by the Drs. Ertl in the USA
[Grateful appreciation is extended to Jan Stokosa CP who kindly provided a bibliography and reprints of most of the articles discussed in this series of commentaries. To read the entire series and to follow all the commentary threads on this topic, click on the June 2003 and subsequent Archives.]
A few years ago, three descendents of Dr. Johann Ertl [all practicing surgeons in the United States at the time] performed a retrospective review of their results from 74 transfemoral revisions and 150 transtibial revisions performed between 1980 and 1995. Since these findings have not been published in any refereed journal to date, the only information currently available is a handout from their lecture at the 1997 meeting of the American Academy Orthopaedic Surgeon's annual meeting. [ Ertl JW, Ertl JJP, Ertl WJ, Stokosa JJ, The Ertl osteomyoplastic lower extremity amputation reconstruction: technique and long-term results, AAOS paper #575, February 17, 1997]
The transtibial series included 143 patients with 150 painful residual limbs that were sufficiently troublesome for the patients to be willing to undergo revision surgery. The average age at revision was 48.5 years. Causes for the original amputations were trauma [63%], vascular disease [28%], infection [7%], and tumor [2%]. [I was interested to note the disproportionately low percentage of dysvascular amputees, since this diagnosis is believed to account for as much as 90% of the new amputations each year in the US.] The average time between the original amputation and the revision surgery was 9.5 years.
Although the decision to operate was based on clinical judgment, these surgeons used a simple evaluation tally sheet to summarize some of the factors that were considered in each case. This "Clinical Assessment Score" subjectively estimated five factors: pain, walking ability, residual limb swelling, duration of daily prosthesis wear, and radiographic presence of a tibiofibular bone bridge. Five points denoted the best possible score while 1 point indicated the worst possible score [e.g. greatest pain, lowest walking ability, and shortest prosthesis wearing times resulted in the lowest numerical score.]
In an effort to summarize these findings, they established an arbitrary "grading" that was simply the mathematical addition of all scores. Greater than 25 total points was considered "excellent" while less than 15 points was "poor", with "good" and "fair" scores falling in between these extremes. [Unfortunately, this scale has never been validated, and there no evidence was offered that each of the five factors were equally significant clinically or that the distinctions between a subscale score of 1 and 2 was identical to the distinction between 4 and 5 or any other one-point difference. Such data would be more accurately expressed as percentage scores for each individual attribute rather than a total additive result..]
Overall, this cohort showed consistently higher numerical values in all five categories at the time of review compared to when first evaluated by their surgeon pre-operatively. Subjective patient satisfaction was recorded at follow-up as "97.3%" indicating that the amputees felt their situation had improved following the surgery. [No information was given on how satisfaction was measured.]
Using the additive "grading" methodology, the authors felt that 73.3% of their results were "excellent and 18.7% were good. "Fair" results were reported for 5.3% and "poor" results for 2.7% of the patients. Pre-operative "grading" data were not reported. Average numerical values for the subscales on pain, walking function, subjective residual limb stability, residual limb swelling, duration of prosthesis wear, and radiographic evidence of a bone bridge were all reported to have increased from the pre-op to the follow-up evaluations, suggesting all these factors were improved.
The only information about the "poor" results in the handout was that those patients all had a dysvascular etiology and reported ongoing post-revision residual limb pain despite having improved scores in the remaining categories. [In other words, approximately 10% of the dysvascular amputees [4 of the 41 individuals] in this series continued to have painful residuums following the Ertl procedure.] The authors concluded that their overall experience in treating amputees had been rewarding.
They also reported their follow-up of 72 transfemoral amputees with an average age of 57.4 years. Cause for the original amputation was again predominantly trauma [60%] and secondarily vascular disease [20%], with smaller percentages due to infection [6%] and tumor [4%].
The Ertl transfemoral operation is very similar to the more familiar transtibial procedure with the exception that the presence of only one bone precludes creating a bone bridge. Instead, the periosteum is used to close the femoral medulary canal in such a way that the end of the shaft is somewhat flared in appearance, in an effort to increase the surface area potentially available for bony end bearing. Once healing is complete, the end of the femur has a contour much like an endoskeletal adapter for a solid-ankle prosthetic foot.
Using the same methodology as in the transtibial series, the authors reported improved average scores in all subscales. The overall results at follow-up were rated by the surgeons as "excellent" or "good" for 67 patients, but "fair" for 2 and "poor" for three. Again, the surgeons noted that the poor results all occurred in patients with peripheral vascular disease. [This means that approximately 20% of the dysvascular transfemoral amputees [3 of 14 individuals] were considered to have poor results following the Ertl procedure.] The authors concluded that the transfemoral Ertl procedure resulted in a high level of patient satisfaction.