Reconstructive ACL surgery: Which graft should you use?

Posted on July 13, 2018

(Last Updated On: July 13, 2018)

A quick primer on what ACL surgery is: Once the ACL is torn it does not repair itself. Thus, a graft is used to form a new ligament. This graft is put in place where the ACL should be.

If you decide to have ACL surgery, one of the next steps is to decide which graft to use.  There are a quite a few options:

  • Patellar graft (click for video) –This is where part of the patellar tendon is cut off and then put back together.
  • Cadaver graft (allograft) –Part of a dead person’s tendon is used. It’s not guaranteed which tendon you will get. You might get an achilles tendon, patellar, anterior tibialis, etc.
  • Achilles tendon
  • Iliotibial band (IT Band)

The last two, an achilles and IT band graft, are very rare in the reconstructive ACL scene. They are no where near as well studied as the patellar, hamstring and allografts. I consider entertaining the achilles or IT band grafts as entertaining the idea of being a science experiment. Because I have no desire to be a guinea pig, I’m not going to discuss these options any further. The patellar, hamstring and allografts will be looked at in depth.

Much like trying to decide whether surgery is a viable option, the research on this area is all over the place. A lot of the studies are crappy retrospective studies, and there aren’t many good, randomized controlled studies. Again, I won’t completely discount the retrospective studies, but I won’t hang my hat on them either.

Autograft versus allograft (a graft from your own body versus a cadaver graft)

The biggest proposed perk of a cadaver graft is you don’t have to recover from the ACL surgery AND another surgical incision. A patellar or hamstring graft from your own body involves cutting open other parts of your body. This is another injury you now have to recover from. However, this doesn’t automatically mean you should use an allograft. Issues arise with them as well.

There are a couple of very good, and recent, randomized studies comparing patellar tendon autografts versus allografts, and hamstring autografts versus allografts. The studies compared aspects like failure rate, stability, activity score, vertical jump, etc.

The autografts consistently come back with better results, but they are very small differences. Differences not worth worrying about.

One thing worth worrying about is the risk of infection between the two grafts. With allografts you really never know for sure what you are getting. The tissue is from another dead person’s body, running the risk of carrying disease. Again, the differences are small, 1.3%, but the autografts group had a ZERO chance of infection. (The cadaver groups had a 1.3% risk.)

Infection can severely hinder the rehab process. Not to mention there’s the (small) chance of amputating your leg, or death. You already have a risk of infection from being in a hospital and having surgery; using an allograft increases that risk.

You aren’t guaranteed to get a patellar or hamstring allograft either. Many times an achilles or anterior tibialis tendon are used. These grafts don’t have as high of a success rate as patellar and hamstring grafts. I haven’t looked into it, but I’m not really sure if you can request the type of allograft you get. Because your name has to be added to a donor list I’m assuming the answer is no. You are likely stuck with whatever is available.

Next, you don’t know the age of the person who is providing the graft. More and more research is coming out suggesting the age of the graft is very important regarding graft failure i.e. older people’s grafts fail more often. I highly doubt you are able to request a “25 year old, male, fully disease free, patellar or hamstring graft.” I could be wrong though.

Keeping with age, a few studies have recently found that, when stratified for age, the rate of graft failure goes way up in younger populations using allografts. Now these studies used things like an achilles graft or an anterior tibialis tendon, which are not ideal, but they are some of the very few that have controlled for age when looking at graft failure rates.

The numbers in these studies are alarming. Failure rates of 25-50% were found in young populations. I really think there is something else going on with these studies -the failure rates are just too high and too different than other studies- regardless, you aren’t going to find this high of a failure rate in autografts, ever. The surgeons I talked to all cited my age as one reason to NOT go with a cadaver graft.

Allografts just have a lot more potential complications. There are even crazy ass stories of donation places lying about the tissue they are giving in order to up sales.

There is only one positive to them: Allografts definitely do have the positive of lessening the damage to the body. Because the autografts involve cutting tissue from other areas of the body there are risks of losing strength at those sites. For example, patellar tendon patients are notorious for having trouble regaining quadriceps strength and knee extension. (This makes sense since a third of their tendon is gone!) It is also common for these patients to have long-term knee pain. For hamstring patients, there is some evidence they have issues regaining knee flexion strength.

Proponents of allografts will say this enables allograft patients to return to their sport quicker.  Maybe an allograft will allow you to return to walking quicker. In regards to getting back to high level sports though, it’s just not true.

Since allografts are from another person’s body they can take longer to remodel into a ligament compared to autografts. Your body is going to have an easier time with its own tissue than someone else’s.

Also, the injuries from an autograft take around 8-10 weeks to heal. The first 8-10 weeks of physical therapy aren’t heavy strength or range of motion wise to begin with.  (You don’t want to stretch the graft out.) Some people aren’t even walking til a couple weeks after surgery. Thus, you’re not losing much by not being able to strengthen those areas yet.

Lastly, look at the people who have the most incentive to return to sports quickly. People where every week matters: professional athletes. Autografts are way more common in this population. If allografts returned people that much faster, those with millions of dollars on the line would use them.

The projected return to sports in an accelerated patellar graft patient is 4 to 5 months. NOBODY with an allograft is returning faster than that. NO.BUD.EE.

I just don’t see much of an argument to get an allograft. If you’re older, like 50, and you don’t care about returning to a high level of activity, then maybe. But if that’s the case, do you even need the surgery to begin with?

Patellar versus hamstring graft 

Which of the autografts should we choose though?

This is actually very similar to examining allografts versus autografts. In terms of the big outcomes like return to activity, stability scores, failure rate, etc. there is minimal differences between the two grafts. Much like allografts and autografts, differences not worth worrying about.

The difference that ends up being pronounced is the potential for complications: Patellar grafts have a greater chance of issues:

  • ACL patients are notorious for having issues with quadriceps strength and regaining full knee extension range of motion. Regardless of the graft used. Taking off a chunk of the patella further adds to the likelihood of these issues.
  • Patellar patients consistently report a greater amount of knee pain post-op. Whether it’s 6 months out or 6 years. Especially when it comes to kneeling and deep squatting.
  • Patellar patients run a greater chance of arthritis down the road due to the aforementioned reasons.
  • The risk is very small, but there is a risk of fracturing the patella when obtaining the graft due to power drilling the patella.

Hamstring grafts don’t run the above potential issues.

There are some studies that find patients have trouble recovering hamstring strength, but there are also studies that don’t find this. Probably indicative of how good a rehab the patient had.  Also, it makes sense hamstring patients will have an easier time recovering knee flexion strength than patellar patients will have recovering knee extension strength. The hamstring will regrow, like a lizard’s tail; this appears much more iffy with the patellar tendon. (If it happens, it can take longer than the hamstring.) Either way, it’s not consistent enough of a finding for me to worry about it.

The meniscus 

A meniscus tear accompanies an ACL tear in around 75% of all ACL tears. If the meniscus is repaired, knee flexion strengthening and range of motion is limited the first 8-10 weeks of therapy. When else is knee flexion strengthening and range of motion limited the first 8-10 weeks? In a hamstring graft patient.

If you are having your meniscus repaired it makes sense to choose a graft that most accompanies your therapy. If you had your meniscus repaired, but chose a patellar graft, you are not only running the risk of limited knee flexion, but having issues with knee extension as well. Rather than have potential issues with my quads and hamstrings, I’d prefer to be able to at least strengthen one area of my leg.

Of course, not all meniscus tears are repaired and not all ACL tears have a concomitant meniscus tear. So this may not factor in.

Adding to the case for a hamstring graft, all the surgeons I’ve talked to prefer the hamstring graft.

The only area I see patellar grafts win out is an earlier return to high level sports. I’m not sure if this true for healing reasons or because patellar grafts have been more common in the past (maybe the therapy is a little more understood, thus, better).

Some people propose the patellar graft heals quicker than hamstring grafts because patellar grafts are attached bone to bone where as hamstring grafts are attached tendon to bone. Some believe bone attached to bone heals quicker than tendon to bone. It’s logical, but unproven. This also assumes no associated meniscus issues. Not typically the case in an ACL tear. Note that if you have a meniscus tear your recovery is going to be longer regardless. Negating the possible quicker return by the patellar graft.

Keep in mind we’re only talking about, maybe, a month difference between the grafts. I’m not worried about whether I play dodgeball in 6 months or 7 months. That’s for people who get paid to play sports. I’m worried about playing sports 10 years from now.

Others will say the patellar graft has greater stability outcomes. It often does. But the patellar graft also runs the risk of losing knee extension…If you have a less mobile knee no wonder it’s more stable…Furthermore, there’s no research to suggest this slightly better stability outcome improves functional outcomes: Both graft patients return to the same level of activity.

-> The one exception here is very high level sports. By definition, 99% of us are not worried about this, but patellar grafts have some advantages.

Summing up

Since that was quite a bit of text I threw together a simple chart marking which graft is better for which outcome:







Return to high level of activity









Range of motion



Associated meniscus repair



Speed of recovery






Recovery of strength


With that said, the overall difference between the grafts are minimal. While all the surgeons I talked to prefer a hamstring graft, it’s not as if they were against the patellar. (From the surgeries I’ve seen, a hamstring graft is easier to harvest.)

One of the biggest factors when deciding which graft might be your preexisting history. If you already have a history of knee pain do you want to choose the graft (patellar) that has the greatest chance of increasing that knee pain? No.

Also of great importance is your surgeon’s preference. If you feel really comfortable with a surgeon, something that is hard to come by, I’d probably just defer to their preference. (As long as you are deciding between a patellar and hamstring!) Knowing someone is going to do a good job likely outweighs the small differences between the grafts. If the guy cutting my leg open prefers the graft that doesn’t involve possibly fracturing my patella, I’m going to go with that.

I’ll end this with saying you need to look at the last ten years on this subject. Up until maybe around 2004, patellar grafts were considered the gold standard. They seemed to have better outcomes, and were way more common. Since there have been changes to the hamstring graft, where the superiority of patellar grafts doesn’t appear to hold true any more. There appears to be a movement towards preferring hamstring grafts.

Some references

Note this is hardly complete, but I’ve nearly gone blind trying to read all this crap.

I tried to post either some of the most important studies or the best ones. One can easily cherry pick studies to make their argument; hopefully I am not doing that.

Remember, when it comes to areas that have a great amount of research behind them there is always one study contradicting another study. Without reading every single study it’s very hard to know what is more prominent, which is often what’s most important. Remember my thoughts above are mostly related to either the few randomized studies, and or, the most apparent trends in the research.

Lastly, remember the date of the research is important when it comes to patellar grafts versus hamstring grafts. We’re most concerned with the most recent research.


-Allograft versus autograft

Age of donor is important:


Higher failure rate in younger athletes:


Higher risk of infection; minimal differences elsewhere

-Patellar versus hamstring 

Greater knee pain and or chances of arthritis in patellar group; minimal differences elsewhere: 


No differences


Potentially quicker recovery in patellar group, length of recovery with meniscus repair:

 Recent advances in the rehabilitation of anterior cruciate ligament 

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.