Letter to the Editor Re: “Dental Implant Failure and the Association with Proton Pump Inhibitors (PPIs) and Selective Serotonin Reuptake Inhibitors (SSRIs)”

by Zeeshan Sheikh Dip.Dh, BDS, MSc, PhD, M.Perio, FRCDC; Aditya Patel DDS, MPerio, FRCDC; Eraldo Batista DDS, MSc (Perio), DSc (Oral Bio), FRCDC

Letter to the Editor:

As a practitioner, who spends a good deal of time embedded in Implant Dentistry, both chair side, and in the educational realm, I was curious to see the latest findings in the article, entitled, “Dental implant failure and the association with proton pump inhibitors and selective serotonin reuptake inhibitors.

There is unquestionably growing data, supporting a potential relationship between these groups of medication, and the possibility of heightened implant failure/complications in such patients. This relationship is rooted in their biologic mechanisms which can impact the bone. Nonetheless, it was my feeling that the two cases represented within this paper, seem to be piggybacking on a possible evidence-based phenomenon, rather than focusing on some of the potential more localized, and pronounced etiologic factors that could have been responsible for the issues shown.

One of the greatest challenges within dentistry, and certainly within the implant world is trying to discern what is responsible for varying degrees and types of complications, including failure. When posed the all too commonly asked question, “why do you think this implant failed”, people are quick to use the broad-stroke answer of “multifactorial”. They then begin to ring off all the possible/prototypical items, such as smoking, uncontrolled diabetes, occlusion, cement, lack of maintenance/poor oral hygiene, poor hard and soft tissue quality/quantity, the restorative material/component/design, etc. etc. The reason we are so fond of using this term “multifactorial” is because it is often difficult to pinpoint what may have alone been the source of an implant failing. Moreover, in many cases the patient does not check off any of the aforementioned and we are simply uncertain of the underlying cause, rather we resort to the old …*%it happens.

That said, when examining the two cases which were put forth in this article, I believe there were many overt possible etiologic factors that would point to the reason for implant failure more equivocally, beyond the backdrop of the patient taking these medications, which may seem coincidental.

If we look at the first case in which an implant was placed in the #26 position we see at the second stage exposure, the healing abutment may not be seated. By the time the implant is restored in image 1D there is already some crestal bone loss, and more importantly the restoration is not seated flush to the implant fixture. This issue of the unseated crown continues to be seen all the way through to image 1F, 4 years later when the implant finally fails. The implant is also placed adjacent to an inclined second molar, along with a space at the mesial and an over-contoured crown, all of which can impair a patient’s ability to gain access to cleanse food impaction, which commonly occurs in such scenarios. The most salient issue is the crown not being seated from the beginning which would not only allow for an area of ingress, but it’s possible that the crown was loose, had micromotion or unfavourable forces throughout the entire life of the implant from the point of restoration forward. Any of the above alone, and certainly in combination would point to reasons why this implant may have failed rather than being on the medication in question.

The second case does not actually denote implant failure but displays what most would refer to as mild crestal bone loss – something that can ensue, even in the most favorable conditions. It seems noteworthy that this fixture design is not platform switch which has long been considered a standard of care for mitigating crestal bone loss to varying degrees. To suggest a medication, possibly having an impact on this commonly seen presentation (eg. mild crestal bone loss of an implant) seems a reach, and may even be difficult to go so far as saying, an association or probable association of a medication, could be discerned from these.

I have an incredible respect for all the authors and their contribution to our field both through this publication and at large and appreciate them bringing further awareness to this plausible interaction. Unfortunately, I feel these specific examples that were supposed to aid in empowering a more generalized possible phenomenon in evidence-based implant dentistry were not necessarily the best to do so.

Dr. Philip M. Walton
US Board Certified,
Diplomate of Periodontics


Dr. Walton,
Thank you for the insightful comments. First and foremost, the major purpose of this paper was to raise awareness on the use of SSRIs and PPIs, and to foster more discussion on the topic. No question that your letter is a proof that this aim was fully accomplished. We know the evidence behind SSRI and PPI use and its association with implant failure based on systematic reviews. We agree that failures can also be multifactorial in nature. The cases presented were representative of situations in which no other known factors currently deemed to be associated with implant complications were involved, except for the fact that the patients were on SSRIs or PPIs. Despite the biological plausibility of the association, we can only hypothesize, based on the best available knowledge, that in these cases the use of medications may have had some role in the complications reported. Obviously, we cannot prove it and that was beyond the scope, and certainly not our aim, as you probably know. The comments to the clinical cases presented are solely based on your interpretations, with all the limitations imposed by 2-D imaging, and certainly not taking into account the reports of the skilled clinicians who followed those cases, so we can understand the skepticism coming from your comments.

For Case 1:

  • The healing abutment is seated properly. The system is Biomet 3I, there is platform shifting from fixture to the healing abutment.
  • 1D crestal bone loss- this is normal bone remodeling and is not indicative of disease, unless the crestal bone loss has been progressive which needs to be tracked for 1 year.
  • Crowns in 1D and 1E are seated. They were torqued to the recommended manufacturer torque.
  • 1F crown was loose on presentation. This was the first time the crown was loose. Note that the crestal bone level is the same from Nov 2017 (1D) to Mar 2021 (1F)

Loose crowns can certainly cause peri-implant bone loss; however, this is usually due to biological plaque accumulation around the crest in which case the pattern of bone loss is usually from the crest first, and subsequently leading to vertical bone loss. This pathology is also accompanied by signs of peri-implant mucositis which is clinically observable. In our patient’s case, there was no noted sign of disease until the loose crown. Given that the crestal bone level was similar to 2017, and loss of integration appear to originate from the body of the fixture, our suspicions were drawn to the systemic changes in the body.

For Case 2, you implied that the changes were not related to disease, rather suggesting that it could be related to the type of implant used. It seems that some of the information depicted in the images was missed, so it is important to emphasize some aspects here. First, we disagree that the loss seen in figure 2C is just mild; It is certainly not something that would be expected in a platform-matching implant in state of health. Several threads are exposed on the buccal aspect of the implant; noteworthy, there is an intrabony component, so the exposure of the surface goes beyond what’s seen. It is also important to notice that there was suppuration (2B), and even in a platform-matching implant system with some crestal bone resorption, suppuration is not expected to occur in a state of health. So, seemingly, we are father apart conceptually here.

The whole point of this paper was to present to the readers something else to consider along with other well-known potential risk factors to implant complications. It’s a topic in frank development that will certainly be further explored in a more thorough basis in the future. In that matter, we believe that we fully reached our goal. Thank you for initiating this insightful conversation.

About the Authors

Dr. Zeeshan Sheikh is a Fellow of the Royal College of Dentist of Canada and Clinical Scientist in Periodontics and Assistant Professor at Dalhousie University (departments of Applied Oral Sciences, Dental Clinical Sciences & Biomedical Engineering).

Dr. Aditya Patel is a Fellow of the Royal College of Dentists of Canada and the current President of the Canadian Academy of Periodontology. He works in private practice in Nova Scotia.

Dr. Eraldo L. Batista Jr. is a Fellow of the Royal College of Dentist of Canada and Associate Professor and Head of the Division of Periodontics at Dalhousie University, and Periodontist at the IWK Health Centre.