Response to Albrektsson T et al.: The long-term efficacy of currently used dental implants: A review and proposed criteria of success. Int J Oral Maxillofac Implants 1986;1(1):11-25.
RESULTS OF THE FIRST FOUR YEARS OF THE TORONTO BRÅNEMARK REPLICATION STUDY FAIL TO MEET DR. ZARB’S CRITERIA FOR
Submitted to the International Journal of Oral and Maxillofacial Implants
A recent article in the inaugural issue of the International Journal of Oral and Maxillofacial Implants purports to be a review of dental implants currently used. The authors — T. Albrektsson, G. Zarb, P. Worthington, and A. Eriksson — admit to "a certain degree of bias" [toward the Brånemark System]. In fact, the article itself is a paper read by Dr. Worthington in Vancouver on Tuesday, May 20, 1986, at a marketing meeting sponsored by Nobelpharma USA, Inc., distributors of the Brånemark System. This article not only chronicles the climb up the ladder from scientific research toward applied biotechnology, but also clearly demonstrates the attempts by some to pull the ladder up behind them. This letter, in response to the Albrektsson et al. article, reveals for the first time the results of the first four years of the University of Toronto Brånemark System Replication Study under the direction of Dr. George Zarb. This information has never been published and was only produced pursuant to legal discovery in Core-Vent Corporation’s litigation with Nobelpharma.
I believe it is crucial that the facts be made clear, that obvious misstatements in the published article be corrected, and that you, the clinician, be given the truth about an article that is currently being used by Nobelpharma USA, Inc., primarily as a widely distributed marketing tool. The following is a Letter to the Editor submitted for publication in the International Journal of Oral and Maxillofacial Implants.
TO THE EDITOR:
The first volume of your journal contained an article by Albrektsson, Zarb, Worthington and Eriksson entitled, "The Long Term Efficacy of Currently Used Dental Implants," in which the authors acknowledged "a certain degree of bias" toward the Brånemark System. This was made abundantly clear by the unfair and inaccurate evaluation of the Core-Vent System and the representation of the success of the Brånemark System. Here are a few examples:
1. STATEMENT: "No results from animal experiments with the Core-Vent Implant have been found in the literature. A histological section from a canine implant published in the booklet from the Core-Vent Corporation demonstrates a patchy contact between the alloy and bone tissue."
FACT: #9; The "booklet" referred to is a reprint of my published article (Oral Implantology, Vol. XII, No. 1, 1986) which references research by Lum and Beirne first published in abstract form and later in the J of Oral and Maxillofacial Surgery, June, 1986. The abstract stated: "All specimens demonstrated new bone growth in the core, vents and at the interface without soft tissue interposition." The published article further confirmed that "the bone metal interface involved almost the entire surface of the embedded implant."
2. STATEMENT: The authors acknowledge a published abstract of 100 consecutively inserted Core-Vent Implants by Lubar and Katin, but failed to mention that 98% achieved osseointegration and maintained it for the two-year observation period. The authors chose to give no credence to this report, stating, "Nor were the parameters for implant success properly defined."
FACT: #9; Lubar and Katin clearly defined their parameters for implant success:
Preliminary results appear to show that the Core-Vent Implant achieves a state of osseointegration on a highly consistent basis. It shows osseointegration as evidenced by the absence of a peri-implant radiolucency and by the lack of clinical mobility.
3. STATEMENT: The authors make reference to two of my articles and state, "It has been repeatedly claimed that equally good results will be achieved with the Core-Vent Implant as with the Brånemark Screw Implant."
FACT: The first reference cited was a Letter to the Editor (Journal of Prosthetic Dentistry, Vol 52, No. 1, 1983) which makes no statements regarding the clinical success of either the Core Vent or Brånemark implants. In the second article (Oral Implantology, Vol XII, No. 1, 1986), I contrasted the two-year results of Lubar, who reported 96% success (functioning, osseointegrated implants) with the Brånemark System (Journal of Prosthetic Dentistry, Vol. 50, No. 2, 1983). Although these reports are short-term, the Albrektsson et al. article confirmed that:
Any implant losses after the first one to two years of function seem to be unlikely, provided osseointegration has occurred. [Article reference: Brånemark, P-I: Personal communication, 1986].
It was also noted in my article that the implants in Lubar’s study had been used in the anterior and posterior of both arches, free standing, splinted to other implants, and splinted to natural teeth –– whereas the Brånemark implant were placed only in the symphysis of the edentulous jaw.
4. STATEMENT: The authors propose yet another definition of osseointegration. Throughout the first seventeen years of Brånemark’s research, only one definition of osseointegration was in existence. In the last three years, however, the number of definitions and qualifications of this interfacial phenomenon seems to be increasing in direct proportion to the increase in popularity of the Core-Vent System.
The 1984 definition (JOMS Vol. 42, No. 8) was "contact on the light microscopic level between living bone tissue and implant."
The 1985 definition (Brånemark text, Quintessence Publishing Co.) added that the implant must be load bearing: "A direct structural and functional connection [must exist] between ordered living bone and the surface of a load-carrying implant."
The 1986 definition in this article included the condition that osseointegration "can only be judged after a substantial period of implant service." This is a "Catch 22" definition in view of the authors’ statement that "the Brånemark results...clearly underscore the basic concept of osseointegration as being the major, if not the exclusive, reason for a successful long-term dental implant attachment."
Newly Proposed Criteria for Implant Success
The authors propose five criteria for implant success and maintain that, if their criteria are not met, then "from both legal and humanitarian points of view, "patients should be informed about the investigative character of such treatment." Neither the FDA nor the ADA subscribe to this criteria. The first two –– "immobile when tested clinically" and absence of "peri-implant radiolucency" –– are the same parameters used in the previously discussed Lubar and Katin Core-Vent study. These are the only clinical tests for osseointegration. Two of the remaining three criteria include "absence of persistent...symptoms, such as pain" and an annual rate of bone loss of less than 0.2mm after the first year. Patient comfort and a slow rate of bone loss would be inherent with an osseointegrated implant if osseointegration determines "a successful long-term dental implant attachment" as the authors claim.
Their last criteria proposes "a success rate of 85% at the end of a five-year observation period." An extensive report on the Swedish clinical results is given by the authors, concluding that "recently published figures from the Brånemark group, and in centers outside Sweden, indicate that the implant’s five-year survival rate in the mandible may, in fact, be higher than 95%." No references were given to support this statement. In fact, they reported on the same articles reviewed by Robert James (Implantologist, Vol. 3, No. 3, 1986) in his article subtitled, "Brånemark ‘Fixtures’ Fail to Meet Recommendations of Harvard Implant Consensus Conference." He concluded:
The reports from Sweden...appear to be designed to provide artificially enhanced results which are not consistent, either internally or with the report by Zarb and Symington.
References to the first two years of the University of Toronto Brånemark replication study reported by Zarb and Symington in 1983 in the Journal of Prosthetic Dentistry (Vol. 50, No. 2, 1983) and follow-up results used in the application for ADA provisional approval are conspicuous by their absence in this article. Through the legal process of discovery in the Core-Vent/Nobelpharma litigation, a report entitled, "5 Year Toronto Osseointegration Results," was produced. It is in the form of a letter dated July 30, 1985 from Dr. John F. Cox on University of Toronto letterhead, to Mr. Ulf Wahlstrom of Bofors Biotech, Gothenburg, Sweden (copies available from Core-Vent Corporation).
The Cox letter appears to include all consecutively inserted implants done at the University of Toronto from October 30, 1979 through March 1, 1984 and can, therefore, be used to substantiate the results reported in Zarb’s 1983 article, which infers that it is reporting on the first twelve patients treated at the University of Toronto:
Forty-five patients were selected for treatment in this Toronto study, and 12 of these patients have worn their prostheses for over 2 years. This report reviews our short-term observations on the 12 treated patients.
Zarb’s 1983 article reported that these 12 patients had a total of 69 implants inserted with 59 implants retained, supporting fixed bridges. This is a success rate of 85.5%, which he concluded "certainly endorses the viability of our Swedish colleagues’ claims." This differs markedly from the results that Cox reports for the first consecutive 12 cases treated at the University of Toronto. The Cox report documents the first 12 cases dating from October 30, 1979 through May 21, 1981 with a total of 73 implants inserted and with 56 implant retained, supporting fixed bridges. This is a success rate of 76.7%.
The Toronto Study used osseointegration as its criteria of success. To call an implant osseointegrated requires, according to the 1985 definition stated above, that it be "load-carrying." Therefore, implants that were not restorable because of their angulation and were "put to sleep," have been included in the unsuccessful count in my analysis of Dr. Cox’s report. Furthermore, the inability to deal with malpositioned implants because the heads cannot be bent, is a reflection of the system and therefore relates to its overall success.
In view of the authors’ statement that "any implant losses after the first one to two years of function seem to be unlikely, provided osseointegration has occurred," the difference in the success rate between the Zarb and Cox reports cannot be explained by the fact that the implant have been observed by Cox for two additional years (Zarb report was 1983, Cox report was 1985).
The authors state that "we would wish and emphasize the desirability, in any clinical trial of implants, of the fate of all consecutive implants being faithfully recorded. Lamentably, this has seldom been done." In light of the controversy over James’ conclusions of "artificially enhanced results" with the with the Swedish reports, the Zarb Replication Study becomes pivotal in validating the authors’ claim. It would appear from a review of the Cox report that, regardless of where the series would be started, no consecutive number of patients received the total of 69 implants reported by Zarb. Therefore, "lamentably" the Zarb et al. 1983 article does not add validity to his "Swedish colleagues’ claims."
The first 12 cases reported by Cox for the University of Toronto Brånemark Replication Study demonstrated an implant success rate of 76.7%, which is minimally above the ADA 75% requirement and fails by over 8% to meet the criteria for five-year success of 85% proposed by the authors (which includes Dr. George Zarb, Director of the Toronto study). The Cox report further showed that the success rate for implants inserted in the fourth year (80.7%) was less than for those inserted in the first year (83.3%). In the first four years, a total of 178 implants were inserted with 147 retained for a success of 82.6%. The results improved with the last 94 implants placed in the fifth and sixth years of the study, but had only a 1-2 year follow-up time as of the filing of the 1985 Cox report. This suggests a rather significant learning curve with the surgical protocol of the Brånemark System. The Cox report also documented bridge fractures which occurred in an astounding 60% of the cases placed in the first four years, and screw fractures of 9% in the same period.
Of the 178 implants placed in the first four years of the Replication Study, 93% were placed in the symphysis of the edentulous mandible which is generally accepted to provide the most ideal opportunity for success because of the quality and quantity of bone in that area and the ease of access for implant insertion. The Swedish results in the maxillary edentulous jaws with the non tapping, solid screw design of the Brånemark Implant were reported in this article to be 10% less successful than in the mandible. This can be contrasted with a recent report on the clinical experiences with consecutively placed Core-Vent Implant by Drs. R. Lubar, R. Katin, A. Buchs, J. Zosky, and D. Patrick. Dr. Dan Laskin, Editor of the Journal of Oral and Maxillofacial Surgery, presented their results of 609 implants at the 1986 AAOMS national meeting. Their initial experience ranging up to 4 years, demonstrated an overall success rate of 96.6%, as determined by the achievement and maintenance of osseointegration with no differences in the success rates reported for the mandible (362 implants) or the maxilla (257 implants). This is possibly due to the self-tapping design of the Core-Vent Implant, which assures immediate fixation even in porous bone and the hollow-vented geometry that allows bone ingrowth. These results suggest that the Core-Vent Implant is more user-friendly than the Brånemark System, especially for use in the maxilla.
5. STATEMENT: The concluding "Author Comments" misunderstand the raison d’etre of Core-Vent Corporation’s new screw fixture. Albrektsson and his colleagues refer to the introduction of the Screw-Vent implant as an "overt attempt to copy the Brånemark screw implant," and falsely state that its introduction was the result of "consequential design changes." Yes, the Screw-Vent was designed to replicate the Brånemark implant in material, size and thread configuration and can be inserted by using many of the same instruments. The Screw-Vent Implant was granted FDA marketing approval (510K #K861426/B) upon the demonstration by Core-Vent Corporation that it was "substantially equivalent" to the Biotes Implant which received its FDA 510K approval in 1984. What the authors failed to appreciate is that the Screw-Vent (3.75mm OD) is smaller in diameter than the smallest Core-Vent (4.3mm OD) and, along with our even smaller Micro-Vent Implant (3.25mm OD), furthers the Core-Vent concept of providing the profession with design and dimensional options to meet the anatomical limitations and functional requirements of the patient.
If these authors are sincere in their plea that "the clinical methods advocated by Brånemark...must be continuously challenged in order to define possible refinements," and if they truly acknowledge that the Brånemark System is "criticized for being complicated and expensive," then they should welcome the Screw-Vent fixture. It has a far simpler surgical protocol, is more versatile in its prosthetic applications and is accessible and affordable to a wider segment of the dental profession and, ultimately, a larger number of patients. Instead, this admittedly biased article injects more confusion, uncertainty and polarization into an area of dentistry that offers hope for a better quality of life to millions of dental cripples.
Gerald A. Niznick, DMD, MSD