A tendon taken from somewhere else in your own body. My first choice is the palmaris longus tendon from the same forearm as the injured elbow. If you do not have one on that side, I check the other forearm. If neither forearm has one, I use the gracilis tendon from the hamstring of your landing leg. In nearly every case I use your own tissue, called an autograft, rather than donor tissue from a tissue bank, called an allograft.
Why a graft is needed at all
By the time most throwers need reconstruction, the ulnar collateral ligament has worn down over years rather than snapped cleanly. There is not enough healthy tissue left to reattach. A reconstruction replaces the ligament with a tendon, which is passed through bone tunnels to recreate the path of the original. Over the following months the tendon remodels and takes on the properties of a ligament.
First choice: the palmaris longus
The palmaris longus is a small tendon that runs down the front of the forearm to the wrist. It is expendable. Removing it costs you no grip strength, no wrist function, and no measurable deficit of any kind, which is exactly why it has been the standard graft for this operation since the beginning. Across the published literature, palmaris longus autograft is the most commonly used graft for UCL reconstruction.
I harvest it through three small stab incisions rather than one long one, and without using a tendon stripper, which protects the tendon and leaves minimal scarring.
Not everyone has a palmaris longus. It is absent in roughly 12 to 15 percent of people, and that rate varies considerably between ethnic groups. Its absence is not a defect and causes no problem in daily life — most people who lack one never find out until a surgeon looks for it.
You can usually check for it yourself. Pinch your thumb and little finger together and flex your wrist toward you. If the tendon is present, a distinct cord stands up in the middle of the wrist. If nothing appears, it may be absent on that side, though it can also be hidden in a thicker forearm. I confirm it in the office before surgery with you awake and participating, and I mark it on your forearm so it can be found through small incisions on the day of surgery.
If there is no palmaris longus on the injured side
The palmaris is frequently present on one side and absent on the other. So if you do not have one in the forearm of your throwing arm, the next thing I do is check the opposite forearm. Taking the graft from the other arm is a long-established option and costs that arm nothing functionally, exactly as on the throwing side.
If neither forearm has one: the gracilis from your landing leg
The gracilis is a tendon on the inner thigh. Like the palmaris, it can be removed without meaningful loss of function, and it provides a longer and often thicker graft.
Which leg it comes from is a deliberate choice. For a right-handed pitcher, the right leg is the drive leg and the left leg is the landing leg. I harvest from the landing leg, because the drive leg is the one generating the power in the pitch, and I would rather not take tissue from the leg doing that work.
That reasoning is shared by most of the surgeons who operate on professional pitchers. When all 52 Major League Baseball team orthopaedic surgeons were surveyed, 72.5 percent of those who responded harvested the hamstring graft from the landing leg, compared with 27.5 percent from the drive leg, a statistically significant difference. The most commonly cited reason was the belief that the harvested hamstring plays less of a role in the pitcher's ability to generate a forceful pitch.
I want to be straightforward that this is reasoning rather than proof. A subsequent study of 191 professional baseball players found no difference in return-to-sport rate, performance after returning, or subsequent hamstring, lower extremity, or upper extremity injuries between grafts taken from the drive leg and grafts taken from the landing leg. The logic is widely shared and makes sense mechanically, but it has not been shown to change outcomes.
Why I use your own tissue rather than donor tissue
Autograft is overwhelmingly the standard. In a systematic review of 47 studies covering 6,671 reconstructions, 92 percent used autograft and only 2.3 percent used allograft, a rate the authors themselves described as much lower than expected. They also noted that the revision rate after allograft reconstruction appears higher than after autograft, while cautioning that the allograft literature is thin.
In professional baseball the pattern is even more pronounced. Large studies conducted with access to Major League Baseball's own surgical records analyze graft type as palmaris longus autograft versus hamstring autograft, because those are effectively the only options in use at that level.
My own read on why: no surgeon wants to be the one who used donor tissue in a professional pitcher who then failed to make it back. The downside risk to a career, and to the surgeon's reputation, is not worth whatever convenience an allograft offers. That is my opinion rather than a documented finding, but it explains the pattern in the data.
I have used allograft in a handful of cases. It remains an option when a patient has no usable autograft or a specific reason to avoid a second harvest site. But it is not my preference, and it is not what I recommend for a competitive thrower.
The real downside of using your own tissue
Harvesting tissue from your own body means creating a second surgical site, and I would rather you hear this from me than discover it afterward.
Palmaris longus harvest is very well tolerated. The incisions are small and patients rarely report lasting problems.
Gracilis harvest is different. Some patients have residual pain at the inner knee and the back of the thigh that persists well beyond the expected healing period. It is not common, it does not prevent return to sport, and it is not disabling, but it is real and it is the main trade-off of choosing autograft over allograft. If you have no palmaris in either forearm, this is a conversation we will have before surgery rather than after.
Does the graft choice change the outcome?
Not meaningfully, among the autograft options. In a study of 566 professional baseball pitchers who underwent reconstruction, the return-to-play rate was 83.1 percent with palmaris longus and 80.7 percent with gracilis, a difference that was not statistically significant. Cohort studies comparing the two grafts in baseball players at two-year follow-up have reached the same conclusion using patient-reported outcome scores.
So the sequence I follow is not about one graft producing better results than another. It is about using the tissue that costs you the least to give up, in the order that costs you least.
The bottom line
Tommy John surgery replaces the torn ligament with a tendon. I use the palmaris longus from the injured side first, the palmaris from the opposite forearm second, and the gracilis from your landing leg if neither forearm has one. All three are your own tissue, which is what nearly every surgeon operating on throwing athletes uses and what the outcome data supports. The trade-off is a second surgical site, which is well tolerated in the forearm and occasionally causes lasting knee or thigh discomfort when the graft comes from the leg.
References
- Erickson BJ, Chalmers PN, Dugas JR, Bach BR, Nicholson GP, Verma NN, Ahmad CS, Romeo AA. Do Major League Baseball team physicians harvest the semitendinosus from the drive leg or landing leg when performing ulnar collateral ligament reconstruction on elite baseball pitchers? Orthop J Sports Med. 2017;5(7):2325967117713987.
- Erickson BJ, Chalmers PN, D'Angelo J, Ma K, Ahmad CS, Dines JS, Romeo AA. Side of hamstring harvest does not affect performance, return-to-sport rate, or future hamstring injuries after ulnar collateral ligament reconstruction among professional baseball pitchers. Am J Sports Med. 2019.
- Erickson BJ, Chalmers PN, Bush-Joseph CA, Verma NN, Romeo AA. Ulnar collateral ligament reconstruction of the elbow: a systematic review of the literature. Orthop J Sports Med. 2015;3(12):2325967115618914.
- Hones KM, Simcox T, Hao KA, Portnoff B, Buchanan TR, Kamarajugadda S, Kim J, Wright JO, King JJ, Wright TW, Schoch BS, Aibinder WR. Graft choice and techniques used in elbow ulnar collateral ligament reconstruction over the last 20 years: a systematic review and meta-analysis. J Shoulder Elbow Surg. 2024;33(5):1185-1199.
- Griffith TB, Ahmad CS, Gorroochurn P, D'Angelo J, Ciccotti MG, Dines JS, Altchek DW, Camp CL. Comparison of outcomes based on graft type and tunnel configuration for primary ulnar collateral ligament reconstruction in professional baseball pitchers. Am J Sports Med. 2019;47(5):1103-1110.
- Trofa DP, Constant M, Crutchfield CR, Dantzker NJ, Saltzman BM, Lynch TS, Ahmad CS. Return-to-sport outcomes after primary ulnar collateral ligament reconstruction with palmaris versus hamstring tendon grafts: a systematic review. Orthop J Sports Med. 2021;9(11):23259671211055726.
- Soltani AM, Peric M, Francis CS, et al. The variation in the absence of the palmaris longus in a multiethnic population of the United States: an epidemiological study. Plast Surg Int. 2012;2012:282959.
- Mirzayan R, Benvegnu N, Sidell M, Acevedo DC, DeWitt DO, Lowe N, Singh A. Functional outcomes of ulnar collateral ligament reconstruction with a novel double suspensory fixation. J Shoulder Elbow Surg. 2020;29(8):1530-1537.
