Femoral Rotation: Let’s Just Use the Posterior Condyles?

The most commonly used landmark to set femoral rotation is the posterior condyles. They are easy to find and easy to use. There is very little measurement error. There is very little interobserver error using them during surgery. From Femoral Rotation : When the Gold Standard is a bit rubbish you’ll see that there is a relationship between the posterior condyles (PCL, Figure 1) and the flexion extension axis (FEA) of the knee. All good then. Lets just use them and get on with it…


Figure 1: Landmarks (Victor, 2009)

Unfortunately for all of us, that’s not going to work – at least not for every knee.

The simplest argument against the posterior condyles is bone loss. They are frequently affected by osteoarthritis. In pretty much every valgus knee and a lot of varus knees there is posterior condylar cartilage loss. This is pretty easy to account for – just add a couple of millimetres or a degree or two of rotation to one side. But when the bone starts getting eroded it gets harder to judge. (Figure 2)

Slide1Figure 2: Severe bone loss on condyles. PCL is 2.4° externally rotated to SEA.

The next problem is that the posterior condyles are not always perpendicular to the FEA (or the SEA, or the axis of rotation of the patella, or the trochlear groove). There is a high degree of individual anatomical variation (even without arthritic bone loss). The most commonly accepted version of this anatomical variation is lateral femoral condyle hypoplasia which is commonly associated with valgus knee deformity. It leads to a relative internal rotation of the posterior condyles to other landmarks such as the SEA and trochlear groove. Unfortunately it seems that lateral condyle hypoplasia is only one of the anatomical variations. There may be an overall association with valgus alignment[1] but internal and external rotation of the posterior condyles relative to the SEA or the trochlear groove can occur unpredictably in any knee regardless of coronal alignment.[2-5] For example, here is a valgus knee with a prominent (? Hyperplastic) lateral femoral condyle leading to an externally rotated PCL. (Figure 3)Slide1Figure 3: Valgus knee with large lateral femoral condyle.

Whenever we are looking at the consistency of a landmark we need to look at what we are comparing it to. As discussed inFemoral Rotation : When the Gold Standard is a bit rubbish our best bet clinically is a CT scan of the Surgical Epicondylar Axis (SEA), though in laboratory situations comparison to the axis of rotation of the tibia or patella is also valid.

Based on the comparison the SEA, on average the posterior condylar line (PCL) is 2° to 3° internally rotated. Once again the key point is the phrase “on average”. In comparison to the SEA all the studies give quite a wide range, generally around 10 ° to 12° with standard deviations of 2° to 2.7°[6-8, 4]. My own CT studies recorded the PCL as -2.1° (internally rotated), SD 2.0°, range -7.7° to +3.6° in normal knees[9] and -2.3°, SD 2.5°, range -8.7° to +4.1°[10] in osteoarthritic knees.

The only conclusion to draw is that a line across the posterior aspect of the posterior condyles does not have a consistent relationship to the SEA. (One answer to this is to use preoperative 3D CT scans to produce cylinders and spheres to find the centre of rotation of the posterior condyles and use the centres of these spheres as a rotational landmark. This produces a landmark which is externally rotated to the PCL as the radii of the medial and lateral femoral condyles are different. I’ll discuss this approach and kinematic alignment in later articles.)

This leads to the next point. Many surgeons who aim for mechanical axis alignment will externally rotate their component 3° from the PCL (=PC+3). This is to compensate for the average medial proximal tibial angle (MPTA) which is 3° varus. (Figure 4) This is once again relying on an average value, so it doesn’t suit a lot of people. The MPTA varies from a 2° valgus to about 8° varus [11]. Measuring the MPTA and correcting by this angle doesn’t work either – there is an association between increasing MPTA and increasing internal rotation of the PCL – but it isn’t a direct relationship. So blindly following the PC+3 doesn’t make a lot of sense, and yet some implant sets only have this as a rotational option. Again, this is based on average values instead of individual measurements.Slide1Figure 4: Mechanical axis alignment. Creating a tibial perpendicular to the mechanical axis of the tibia. More bone is taken off the lateral tibial plateau than medial, therefore the femur is externally rotated to take more bone off the medial femoral condyle than lateral.

So how do we measure anatomical variations? One approach has been to get a preoperative CT scan or MRI and measure the angle between the PCL and SEA. Then during surgery with conventional instruments use a sizing guide which allows variable angles from the posterior condyles and dial in the measured angle. Michaut et al used this technique to get 77% of cases within 2° of the SEA. This is an improvement over using the PC+3, but still nowhere near perfect. It can be inaccurate on 2D slices so consider using 3D reconstructions (Osirix software is probably the easiest if you can’t get your implant company to do preoperative plans for you). Obviously it requires a scan – which not everyone wants to get. It also doesn’t take into account variations in the anatomy of the trochlear groove relative to the posterior condyles (more on that later).

So, if you like the posterior condyles, be careful – they may be easy to use but they are often wrong. Get a sizing guide you can adjust. Adjust for bone and cartilage loss. Compare you PC+3 to other landmarks, in particular the trochlear groove – which is the next story…


  1. Luyckx T, Zambianchi F, Catani F, Bellemans J, Victor J. Coronal alignment is a predictor of the rotational geometry of the distal femur in the osteo-arthritic knee. Knee Surg Sports Traumatol Arthrosc. 2013;21(10):2331-7. doi:10.1007/s00167-012-2306-x.
  2. Jones C, Nawaz Z, Hassan A, White S, Khaleel A. The variability in the external rotation axis of the distal femur: an MRI-based anatomical study. European Journal of Orthopaedic Surgery & Traumatology. 2016;26(2):199-203. doi:10.1007/s00590-015-1719-x.
  3. Chao TW, Geraghty L, Dimitriou P, Talbot S. Averaging rotational landmarks during total knee arthroplasty reduces component malrotation caused by femoral asymmetry. J Orthop Surg. 2017;12(1):74. doi:10.1186/s13018-017-0575-2.
  4. Thienpont E, Schwab P-E, Paternostre F, Koch P. Rotational alignment of the distal femur: anthropometric measurements with CT-based patient-specific instruments planning show high variability of the posterior condylar angle. Knee Surg Sports Traumatol Arthrosc. 2014;22(12):2995-3002. doi:10.1007/s00167-014-3086-2.
  5. Amaranath JE, Moopanar TR, Sorial RM. Defining distal femoral anatomy for rotational alignment in total knee arthroplasty: a magnetic resonance imaging-based study. ANZ J Surg. 2014;84(11):852-5. doi:10.1111/ans.12708.
  6. Griffin FM, Insall JN, Scuderi GR. The posterior condylar angle in osteoarthritic knees. J Arthroplasty. 1998;13(7):812-5.
  7. Nagamine R, Miura H, Inoue Y, Urabe K, Matsuda S, Okamoto Y et al. Reliability of the anteroposterior axis and the posterior condylar axis for determining rotational alignment of the femoral component in total knee arthroplasty. J Orthop Sci. 1998;3(4):194-8.
  8. Poilvache PL, Insall JN, Scuderi GR, Font-Rodriguez DE. Rotational landmarks and sizing of the distal femur in total knee arthroplasty. Clin Orthop. 1996(331):35-46.
  9. Newman Cea. Femoral Rotational Asymmetry is a common anatomical variant. Submitted 2017. 2017.
  10. Talbot S, Dimitriou P, Radic R, Zordan R, Bartlett J. The sulcus line of the trochlear groove is more accurate than Whiteside’s Line in determining femoral component rotation. Knee Surg Sports Traumatol Arthrosc. 2015;23(11):3306-16. doi:10.1007/s00167-014-3137-8.
  11. Hayasaka Sea. Femoral Rotational Asymmetry is linked to proximal tibial varus. 2017 (submitted).



6 thoughts on “Femoral Rotation: Let’s Just Use the Posterior Condyles?

  1. Hi Simon,
    I routinely use the Whiteside line to reference femoral rotation when I make the femoral cuts in TKR. I do not usually assess rotation pre-op with a CT scan. Do you routinely perform CT as a part of your pre-op work-up or reserve this for patients where you anticipate a problem e.g TKR in a femoral diaphyseal deformity?

    Liked by 1 person

    1. I don’t think that CTs are essential. I use a 3D version of Whitesides line called the Sulcus line and also reference the posterior condyles. I’ll describe it in detail in the next couple of articles. It is at least as accurate as using CTs and averages out the shape of abnormal, asymmetrical femora


  2. Hi Simon

    The problem of femoral rotation can somewhat be solved by using navigation. I get pre operative MRI on all my patients and asses 2 things – 1. Cartilage loss and 2. the PCA to the TEA Angle ( my data average 3.6deg IR)
    I also preoperatively template my MPTA and respect my tibia according to it. ( anatomical ) this then allows me to balance my femur on my tibia ( usually if my TEA registration is correct( hard to know always unless have nipple epicondyles) i m within 1-1.5 deg of my pre op rotational templating) if the native PCA is more then 5 deg iR I would consider closing my natural lateral laxity to avoid PFJ issues. If there is no posterior condyle cartilage loss I will resect the same amount of bone on each condyle and this usually gives my trapizoidal gaps ( ala natural knee – lateral is loose)
    Your blogs are interesting though


    1. Thanks Kosh. I like the amount of thought that you obviously put into each case. A lot of what I’m trying to promote is to encourage people to get their heads around these concepts and treat each cases individually. You are clearly doing a version of kinematic (anatomic) alignment – but making some compromises for individual variations in anatomy. I’m interested in your comment that you will “close down your natural lateral laxity” (by which I assume you mean externally rotate your femur) if the PCL is >5deg to avoid PFJ issues. I like this idea as it fits with the work I’m going to present in the next couple of articles on femoral rotational asymmetry and individual variations in trochlear groove anatomy. Thanks


  3. Hi Simone,

    Thank you for sharing the article. It helped me understand the complexity and challenges surrounded by femoral rotation in a very coherent way. Being a former design engineer, I could see so much value in this article. Hope you will continue sharing your insights with us.

    On selection on reference axis: I do believe PreOp planning can immensely help in order to draw the conclusion on the selection of reference axis for rotation. I reckon reference axis should be chosen based on the bony condition of the particular patient. PCA is the better option except for the cases of hyplasia, as you mentioned. The only challenge I see, sometimes floating osteophytes attachment at the PCL line which can hamper the angular measurement. Regarding TEA, reproducibility of medial epicondyle could be challenging and even navigation may not be perfect in capturing these landmarks precisely.

    On 2D landmark vs 3D landmark: I would like to know your opinion on the accuracy of 2D and 3D landmark. The one problem with 2D scan can be that both landmark medial epicondyle and lateral epicondyle might not exist in the same slice. Also, the orientation of CT co-ordinates may play a major role in measurements. Talking about 3D landmark- The accuracy of the final 3D model depends on the algorithm used and experience of manual segmentor who is performing this work. The extra added manual error can be the difference between 2D & 3D landmark.

    Rotational measurement variation in MRI & CT modality: Assuming we are picking the landmark on a 3D model which is extracted after segmentation process. There would be variation in angle between PCL and TEA in MRI and CT models as MRI model consists cartilage, which may not be uniform on both sides. This variation can be almost of 1 deg. Do you think Preop Planning using MRI modality could be the better option which represents surgical condition better or you will go with CT extracted angulation for setting up the rotation??

    Rotation in extra-articular deformity and severely bow cases: As literature suggests, coronal and sagittal bow of bone affect the post-op coronal and sagittal alignment of the knee. How do you tackle the rotation in EAD and bow cases and how much it can affect the post-op rotational alignment?


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s