Are NSAIDs a Viable Option During Bone Healing?

Point

Yes. These authors contend that NSAIDs inhibit inflammation and provide pain relief, and avoid the side effects of opioids, sharing their research and an approach to a multimodal postop pain control regimen.

By Mindi Dayton, DPM, MHA; Jesseka Kaldenberg-Leppert, DPM, MSA; Paul Dayton, DPM, MS

Pain management following foot and ankle surgery is a primary concern for patients as well as surgeons. Traditional post-surgical pain management relies mainly on narcotic pharmaceuticals and, to a lesser extent, on non-prescription or non-pharmaceutical measures.

For over a decade, two of the authors (MD, PD) have employed a multimodal pain management regimen for their postoperative patients with great success. Adopting this multimodal pain management regimen came from a desire to provide patients with improved pain control while limiting the utilization of potentially addictive narcotics. According to the Centers for Disease Control and Prevention (CDC), more than 70% of drug overdose deaths in 2019 involved an opioid.¹ A multimodal pain management strategy could help to prevent these deaths.

A priority in our rehabilitation philosophy for the majority of our surgical procedures is rapid return to both non-weight-bearing range of motion exercises as well as early weight-bearing to prevent the detrimental effects of prolonged immobilization.

What Does the Authors’ Postop Regimen Entail?

We have found the following postoperative multimodal pain management regimen helps us and helps patients achieve these goals:

• Detailed preoperative education and coaching of patients and their support system
• Preoperative administration of gabapentin 60 minutes prior to the procedure
• Intraoperative administration of ketorolac (Toradol) if not contraindicated
• Preoperative ankle block with 0.25% bupivacaine for preemptive analgesia
• Limited tourniquet times as well as use of lower pressure (225–250mm Hg for thigh and 200–225mm Hg for ankle)
• Meticulous soft tissue and bone handling
• Postoperative field block with 0.25% bupivacaine
• Limited post-surgical bandaging
• Consistent use of a specific combination non-narcotic medications beginning two hours postoperatively, prior to the anesthetic block wearing off. This continues every 8 hours and usually continues for 7–10 days.
• We choose a non-steroidal anti-inflammatory drug (NSAID) (ibuprofen 600 mg), acetaminophen 650 mg, and gabapentin 100 mg taken every 8 hours, all at the same time.
• Prescribe 10 tablets of oxycodone
• 5 mg to be taken as needed for breakthrough pain.

Success with this treatment regimen is likely due to harnessing the synergistic effects of medications by addressing patients pain via different mechanisms.² Pre- and postop anesthetic blocks attenuate pain at the central nervous system level while NSAIDs inhibit inflammation and the pain cascade. In addition, when avoiding opioids, one may also avoid negative side effects such as nausea, vomiting, sedation, constipation, and respiratory depression. Multimodal pain management can also decrease the time to discharge from the post-anesthesia care unit as well as hospital costs.²

Educating patients, as well as their support system, is paramount. Stress the  importance of timing—starting medications immediately even if pain has not begun—and consistency. With correct timing and consistency, one can avoid pain crises. It is important to note that in our patient population, most utilize only 2–3 pills on average throughout their entire surgical recovery, and it is very rare that patients request a refill of their oxycodone prescription. Additionally, if a patient has a medical condition or other reason NSAIDs cannot be used safely, one should exclude NSAIDs from the patient’s regimen; however, it is rare that the opioid prescription or usage is changed.

Reviewing the Research on Postop Pain

NSAIDs are the safest and most effective drugs to treat postoperative pain, with few exceptions. Still, some believe one should generally avoid NSAIDs in orthopedic and podiatric procedures due to concern for delayed bone healing. Is this really true? A systematic review by Marquez-Lara and colleagues attempted to answer this question.³ The authors included 12 clinical articles and 24 literature reviews in their review. They identified a wide range of non-union rates associated with orthopedic procedures in the NSAIDs group (0–65% non-union) compared to the control group (0–19%). The study notes that besides NSAIDs, multiple factors can affect non-union rates, eg, soft tissue and bone handling. Overall, the authors found more clinical studies concluded NSAIDs were safe compared to those studies that recommended NSAID avoidance. They noted a significant lack of prospective randomized control trials regarding NSAIDs and their effects on bone healing; many more trials are needed in the future.³

Along with colleagues, we have published several articles related to the outcomes of arthrodesis procedures of the foot. The following peer-reviewed publications highlight the healing outcomes of patients who underwent bone arthrodesis. We included the multimodal pain management regimen we described in all the patients in these studies and included the routine use of NSAIDs if no medical contraindications existed.

In a recent peer-reviewed publication, we reported on the outcomes of 60 patients that underwent a triplane first MTPJ arthrodesis for hallux valgus, at a mean 28.4 months postop.⁴ The patients were consecutive patients who met inclusion criteria and were willing to complete a phone survey on their subjective outcome. A certified independent radiologist reviewed the radiographs and identified no non-unions. Additionally, patient satisfaction was extremely high, with 46% (28/60) noting that their recovery was easier than expected and 41.7% (25/60) noting that the recovery was as expected.⁴

In 2020 we reported on 108 patients (109 feet) who underwent triplane tarsometatarsal arthrodesis for hallux valgus between 2014 and 2017, with a mean follow up of 17.4 month. Our results revealed a 100% union rate and a 1% recurrence rate (1/109).⁵

Dayton and colleagues reported on the healing of 195 first ray arthrodesis procedures with biplanar plating and early protected weight-bearing in a 2019 publication.⁶ Eighty-five feet underwent first MTPJ arthrodesis, and 110 feet underwent first tarsometatarsal arthrodesis (TMT). At the final radiographic follow-up, 97.44% of all cases had shown progressive osseous gap filling at the arthrodesis site, a stable position of the bone segments, and intact hardware without loosening, 98.24% of the first MTPJ arthrodesis group and 96.82% of the first TMT arthrodesis group. Five (5.43%) feet had the presence of lucency at the fusion interface at the final follow-up, without positional change or hardware failure. Four (1.8%) feet had a failure of the hardware, loss of position, or frank gapping at the fusion site.⁶

A 2004 Dayton and McCall study reported on a series of 42 patients with 47 first MTPJ fusions with early weight-bearing.⁷ They reported a 100% union rate based on clinical and radiographic findings. This was published prior to our consistent use and education of patients on taking the postoperative pain medications noted in the protocol. However, patients were not told not to take NSAIDs, and they were actively encouraged to do so, especially those with pain not controlled by narcotics, as prior to 2010, narcotics were our primary source of pain control for patients postoperatively.

Two authors (MD, PD) are co-principal investigators in a prospective, multicenter, study designed to evaluate outcomes of bunion correction with a specific instrumented technique. They are also co-authors in the recent publication related to this study, which reported on the one- and two-year analysis assessing radiographic and patient-reported outcomes.⁸ At two-year follow-up, 1/117 (0.9%) had a reported non-union. A subset of patients in this group were patients of the authors (MD, PD), and all utilized the multimodal pain management regimen described.

Many things contribute to achieving a union at an arthrodesis and/or osteotomy site, including, but not limited to:

• proper patient and procedure selection;
• surgical technique including fusion site preparation with complete removal of subchondral plate;
• avoidance of heat from surgical instruments at the surgical site which can cause bone necrosis;
• full-thickness dissection, which preserves soft tissue and periosteal blood supply;
• multiplanar stability provided by fixation; early active rehabilitation, which promotes a more rapid recovery;
• and patient selection based on health and comorbidities.

In Conclusion

In our experience, multimodal pain management with NSAIDs is an excellent approach to treat postoperative pain and we have not identified any complications with respect to healing as evidenced by our published outcomes as well as the work of others.

Mindi Dayton, DPM, MHA, FACFAS, practices at Foot & Ankle Center of Iowa in Ankeny, Iowa. Dr. Dayton is board-certified by the American Board of Foot and Ankle Surgery in foot surgery and reconstructive rearfoot/ankle surgery. She is a fellow of the American College of Foot and Ankle Surgeons.

Jesseka Kaldenberg-Leppert, DPM, MSA, AACFAS, practices at Foot and Ankle Center of Iowa.

Paul Dayton, DPM, MS, FACFAS, practices at Foot & Ankle Center of Iowa in Ankeny, Iowa. Dr. Dayton is board-certified by the American Board of Foot and Ankle Surgery in foot surgery and reconstructive rearfoot/ankle surgery, and is also a fellow of the American College of Foot and Ankle Surgeons.

Counterpoint

No. This author argues the excessive use of NSAIDs inhibits the formation of new bone, and recommends exercising caution when developing a pain management protocol.  

By Marlene Reid, DPM

As I enter my fourth decade of private practice, I have come to understand what my older colleagues meant when they said medicine is cyclic: Techniques or theories fall in and out of favor over time. For podiatric medicine and surgery, the use of implants, metatarsal osteotomies, and joint fusions certainly follows that definition. The use of non-steroidal anti-inflammatory drugs (NSAIDs) for surgical patients does as well.

I recall many DPMs older than me using NSAIDs preoperatively to control postop swelling and pain. Similarly, many in my generation trained using a full dose of steroid injected into a surgical site intraoperatively. Over the next 20 or so years, the use of both NSAIDs, either preoperatively or postoperatively, and steroid injections became unpopular due to their inhibitory effect on healing.

It seems as though the pendulum has swung again, and the newer group of podiatric surgeons favor the use of NSAIDs postoperatively. I’ve observed many contemporary colleagues stating that either they use NSAIDs after bunionectomies as long as needed or always use them unless there is a medical contraindication. In fact, in an online discussion I participated in, nearly ¾ of the participants felt this way. On the other hand, it seems I notice there are others that do not use NSAIDs postoperatively, or only do so for a few days. I fall into the category of only using NSAIDs for a few days.

Why the Author Avoids NSAIDs During Bone Healing

While NSAIDs clearly reduce pain and inflammation, my reason for using ketorolac 10mg for 2 or 3 days postop is to reduce the number of opioids my surgical patients require. In today’s climate of opioid abuse, I understand why many podiatric surgeons turn to NSAIDs for pain control. However, while serving as a guest editor and author on research regarding regenerative medicine, I affirmed my belief that the risk of overly inhibiting the inflammatory process leading to a chronic state is real, and one should avoid it in both soft tissue and bone healing. Intuitively, years ago, I began to limit my intraoperative steroid use to literally a drop of short-acting steroid directly on the bone prior to closure.

As you will recall, the initial two stages of healing require inflammatory cells to progress from the acute phase of healing to the subacute phase of healing. Cellular infiltration is necessary to allow degradation and phagocytosis, all leading to the second phase of angiogenesis and fibroblast production of collagen. It is also significant to note that the timeframe for healing varies widely, and this process can take anywhere from a few days to a few weeks. When this process stalls, the transition to the remodeling phase may be affected.¹

In my experience, today’s regenerative medicine techniques such as shockwave, platelet-rich plasma (PRP), and radiofrequency enable the “stalled” environment to return to an inflammatory one to allow healing to begin again to enable the completion of all three phases. It is interesting to note that the remodeling phase has long been called “chronic inflammation.” Certainly, we now know that that term is an oxymoron, but it explains why our previous (lack of) understanding of this phase was that it extended from 2 to 3 weeks to years to resolve.  

New bone forms mainly by endochondral ossification, in which cartilage formation precedes bone.² Remodeling of healing bone includes the ability to regenerate shape, strength, and function, and in addition to the sequence of events described above, also depends on the stability of the fracture or osteotomy.³ I believe excessive use of NSAIDs may inhibit this process.

A Closer Look at the Research

I believe the research supports both my intuition and my reasonings.  

Quan and colleagues found that NSAID use, among other factors studied, contributed to non-unions in limb fractures.⁴ A retrospective study compared 223 patients with non-unions to a control group of 446 patients with successfully healed fractures.⁴ Authors used a multivariate logistic regression model to evaluate relevant factors for the non-union cases. The odds ratio estimates placed NSAIDs third in a list of multiple factors that included delayed weight-bearing, failed internal fixation, and infection, only falling behind osteoporosis and open fractures. Interestingly, smoking and a medical history of diabetes did not present as a factor in this study, nor did age, sex, hypertension, or alcohol use.⁴

Al Farji and colleagues conducted a 2021 review of six human randomized controlled trials (RCTs) through October 2018 that compared the occurrence of non-union in patients who received NSAIDs to control groups.³ While acknowledging NSAIDs’ effect on bone healing is debatable in human and animal studies, the authors sought to evaluate only RCTs to provide a greater understanding of the true effect of NSAIDs on human bone healing. The most significant part of this meta-analysis evaluation was the specifics the authors found on the length of time patients utilized NSAIDs. Their findings showed those who received NSAIDs for a short period (less than 2 weeks) did not show any significant risk of non-union compared to those who received NSAIDs for a long period (greater than 4 weeks).  This certainly supports the need to allow progression from the subacute phase of healing to the regenerative phase of healing.

Al-Waeli and coworkers also acknowledged the contradictory results in human studies; however, they found the research of animal trials were more consistent in suggesting that decreased bone healing was associated with NSAID administration.⁵ Moreover, the authors found there was no meta-analysis study to date prior to their review in 2021. The authors identified 32 studies that compared the effect of administration of one or more NSAID on the biomechanical characteristics maximum force to break, stiffness, and work-to-failure to a control group using micro-computed tomography (μ-CT), and histomorphometric measurements which are widely used in animal bone studies.  

Thirty of these studies found the maximum force to fracture was significantly decreased in those animals that had NSAIDs, as well as an overall decrease in bone stiffness and work-to-failure properties as compared to control groups.⁵ Normal healing of bone is often measured by the ease of re-fracture and these studies support bone healing delay with the use of NSAIDs.
It is interesting to note that this review also compared mice with other animals, the genders of the animals, and the ages of the animals. The authors did not find a statistically significant between any of these comparison factors.⁵

NSAIDs affect the inflammatory response by inhibiting cyclooxygenase (COX) enzyme activity, which decreases the synthesis of prostanoids, mediators of the inflammatory response. The two types of COX enzymes are divided based on the specific role they play in the pathway. The role of COX-1 is the synthesis of prostaglandins under normal physiological conditions, whereas COX-2 is synthesized under an inflammatory state and induces the release of prostaglandins,³ which are needed for cellular recruitment.⁶ Multiple studies have shown the use of COX-2 selective NSAIDs is detrimental to healing,^6,7 including the need of COX-2 for endochondral ossification and osteoblast recruitment.²

A 2021 study of mouse pre-osteoblastic cells exposed to NSAIDs while comparing COX-2 selective inhibitors to non-selective COX inhibitors as well as steroidal prednisolone.² The authors found all NSAIDs tested did show a negative impact on pre-osteoblast cell growth.  

However, this same study focused on osteogenic differentiation by way of measuring alkaline phosphatase (ALP) activity and calcium deposition.² The authors’ findings were drug-specific and lacked consistency in determining if COX-2 differed from non-selective COX inhibitors. Interestingly, the study found prednisolone to increase ALP activity but decrease calcium deposits.    

For me, the most significant research I found is that of Al Farji and colleagues, who independently evaluated the effects of NSAIDs on the length of time used.³ Most bunionectomies require modulation of postoperative pain for less than four days. I intend to maintain my current protocol of using ketorolac 10mg for no more than three days, to hopefully balance its benefits with the potential downsides. Overall, I believe one should exercise caution and not underestimate the real impact of NSAIDs on bone healing. 

Dr. Reid is the current Secretary-Treasurer of the Federation of Podiatric Medical Boards. She is a Past President of the Illinois Podiatric Medical Association and the American Association for Women Podiatrists. Dr. Reid is a Fellow of the American College of Foot and Ankle Surgeons and a Distinguished Fellow of the National Academy of Practitioners. She is a Delegate to the American Podiatric Medical Association House of Delegates and is in private practice in Naperville, Ill.

Point References
1.     Drug Overdose Deaths. Centers for Disease Control and Prevention. https://www.cdc.gov/drugoverdose/deaths/index.html. Accessed July 25, 2022.
2.    Jin F, Chung F. Multimodal analgesia for postoperative pain control. J Clin Anesth. 2001; 13(7):524–539.
3.    Marquez-Lara A, Hutchinson I, Nuñez F Jr., Smith TL, Miller AN. Nonsteroidal anti-inflammatory drugs and bone-healing: A systematic review of research quality. JBJS Reviews. 2016; 4(3):e4. doi: 10.2106/JBJS.RVW.O.00055
4.    Dayton M, Dayton P, Topher C, Thompson J. What do patients report regarding their real world function following first metatarsophalangeal arthrodesis for hallux valgus. J Foot Ankle Surg. Published online July 19, 2022 doi:10.1053/j.jfas.2022.07.003
5.    Dayton P, Carvalho S, Egdorf R, Dayton M. Comparison of radiographic measurements before and after triplane tarsometatarsal arthrodesis for hallux valgus. J Foot Ankle Surg. 2020 Mar-Apr;59(2):291–297. doi: 10.1053/j.jfas.2019.08.020. PMID: 32130993.
6.    Dayton P, Santrock R, Kauwe M, Gansen G, Harper S, Cifaldi A, Egdorf R, Eisenschink J. Progression of healing on serial radiographs following first ray arthrodesis in the foot using a biplanar plating technique without compression. J Foot Ankle Surg. 2019 May;58(3):427-433. doi: 10.1053/j.jfas.2018.09.001. Epub 2019 Feb 23. PMID: 30803912.
7.    Dayton P, McCall A. Early weightbearing after first metatarsophalangeal joint arthrodesis: a retrospective observational case analysis. J Foot Ankle Surg. 2004 May-Jun;43(3):156-9. doi: 10.1053/j.jfas.2004.03.007. PMID: 15181431.
8.    Liu GT, Chhabra A, Dayton MJ, et al. One- and two-year analysis of a five-year prospective multicenter study assessing radiographic and patient-reported outcomes following triplanar first tarsometatarsal arthrodesis with early weightbearing for symptomatic hallux valgus. J Foot Ankle Surg. 2022 Apr 27:S1067-2516(22)00118-1. doi: 10.1053/j.jfas.2022.04.008.

Counterpoint References
1.    Sheen JR, Garla VV. Fracture healing overview. StatPearls [Internet]. 2022. Available at: https://www.ncbi.nlm.nih.gov/books/NBK551678/ . Accessed September 7, 2022.
2.    Hadjicharalambous C, Alpantaki K, Chatzinikolaidou M. Effects of NSAIDs on pre-osteoblast viability and osteogenic differentiation. Exp Ther Med. 2021;22(1):740.
3.     Al Farii H, Farahdel L, Frazer A, Salimi A, Bernstein M. The effect of NSAIDs on postfracture bone healing: a meta-analysis of randomized controlled trials. OTA Int. 2021;4(2):e092.
4.    Quan K, Xu Q, Zhu M, Liu X, Dai M. Analysis of risk factors for non-union after surgery for limb fractures: A case-control study of 669 subjects. Front Surg. 2021;8:754150.
5.    Al-Waeli H, Reboucas AP, Mansour A, Morris M, Tamimi F, Nicolau B. Non-steroidal anti-inflammatory drugs and bone healing in animal models-a systematic review and meta-analysis. Syst Rev. 2021;10(1):201.
6.    Kigera JWM, Gichangi PB, Abdelmalek AKM, Ogeng’o JA. Age related effects of selective and non-selective COX-2 inhibitors on bone healing. J Clin Orthop Trauma. 2022;25:101763.
7.    White AE, Henry JK, Dziadosz D. The Effect of nonsteroidal anti-inflammatory drugs and selective COX-2 inhibitors on bone healing. HSS Journal. 2021;17(2):231-234.