Whiteside’s Line: Throw it out and start again

Posted by

Whiteside’s Line: Throw it out and start again

This headline may sound a bit harsh, and no offense intended to Dr Whiteside, but I’ll show you why I think we can do better. I’ll show you why we need to do this and then provide the published evidence.

The anteroposterior axis (APA, also known as Whiteside’s Line) has been around since the 1990s[1]. (Figure 1) It has been shown on multiple studies to be highly inaccurate and difficult to reproduce.[2-6] These studies report ranges for WL compared to the Surgical Epicondylar Axis (SEA) of between 15° and 22° with standard deviations between 4.2°[5] and 7.6°.[2]


Figure 1. Whitesides Line, also known as the anteroposterior axis (APA), is a 2D line which runs from the centre of the intercondylar notch to the deepest point of the trochlear groove anteriorly.

At this point I think that we want to decide what it is that we are really trying to achieve. I feel that what we want is to isolate the rotational component of the alignment of the trochlear groove. It turns out that the APA uses the wrong anatomical points on the femur and is unable to correct for variations in the 3D anatomy of the groove. We want something that is more reproducible and which can be used both on preoperative scans and during surgery. So let’s start from scratch and think about how we should go about it.

  • Mark out the whole groove (not just two points). Place points along the deepest part of the floor of the groove starting from the intercondylar and heading anteriorly. If you do this you will very quickly find that the two points used in APA are both not great choices. Posteriorly the deepest part of the trochlear groove is almost always lateral to the centre of the intercondylar notch.[7] Anteriorly the proximal section of the trochlear groove is unreliable. It is often affected by arthritic damage, but even in normal knees the deepest point of the groove frequently veers off medially or laterally in the last couple of centimeters. (Figure 2) Victor et al found that the proximal point of APA was highly variably and poorly reproducible.[8] The reliable section runs up from the notch and stops a a centimeter or two before the proximal extent of the chondral surface. Those of us doing knee replacements will be saying that this is what we’ve been doing for years, drawing in the deepest part of the groove with a diathermy. And you’d be right, but it’s the next steps that make it more accurate. I’ll show you how you can improve on what you are currently doing.


Figure 2. 3DCT reconstruction. The proximal section of the trochlear veers off from the vertical section. The posterior point is lateral to the centre of the intercondylar notch.


  • We’ve drawn in a curve, not a straight line. (Figure 3)

SL curved

Figure 3. The trochlear groove is a curve.


It has a rotational (axial) component, but it also has a coronal component. That is, it runs in a coronal direction relative to the mechanical axis of the femur. Unfortunately every trochlear groove runs in a different coronal direction. The amount of individual coronal variation runs from at least 9.4° varus to 7.3° valgus to the mechanical axis[5]. (Figure 4).


Figure 4. A-P view of the femur. The trochlear groove has a variable coronal alignment.

It turns out that this is very important and we need to correct for it. If you want to isolate the rotational component of the curve you need to be looking directly along coronal direction in which it is running. If you don’t do this you will get a large geometric error called a parallax error. It will frequently change your rotational angle by 5-10° and you won’t even notice it happening.

Here is a video of a femur with the trochlear groove marked in.

Figure 5. Video of trochlear groove.

You can see how it is a curve when viewed from most angles but it becomes a straight line when we are looking directly along the line of the curve. By drawing in the whole curve, we are able to find the coronal direction of the groove so we can tell the direction we need to be looking to isolate the rotational component. You can’t do this with the two points of APA.

Have a look at the Figures 6 and 7. I have marked in the two points used to derive the APA, and the two points of the SEA. The points are identical in each picture. The angle between them changes depending on the direction in which you look at the end of the femur. The amount of change in rotational angle is huge. I have also put in the video so you can see what is happening (Figure 8).

Slide2 Slide1

Figure 6 and 7. The same femur and same anterior and posterior points. The different angles are purely caused by parallax error.

Figure 8. Video of the APA angle changing with the direction we look at the femur.

We’ve now got a line that uses different points to APA – we don’t use either the centre of the intercondylar notch or the most proximal part of the trochlear groove proximally. We then correct for variations in the coronal alignment of the groove when we measure its rotation – something that is impossible with APA’s two points. Therefore we have a different landmark – which we have termed the Sulcus Line of the trochlear groove (SL).

Enough theory, now for the evidence.

We analysed a series of 3D CT scans. We measured the Sulcus Line (SL) looking along the coronal alignment of the groove in each case. We measured APA between two points looking along the mechanical axis of the femur in each case. The results (Figure 9) show a much lower range of measurements using the SL technique compared to the APA technique (9.6° compared to 19.6°). Our APA results were very similar to those from other APA studies. This was published in KSSTA 2015[5].


We have published studies on using this technique in a cadaver lab and a clinical trial, both of which confirmed that the SL is more accurate and reproducible than the APA.[9, 10] Using the SL requires the use of an instrument we have designed more on that later in Femoral component rotation: Find your groove more accurately, and here’s the link to the instrument (Enztec SL instrument). Most of the advantage is in the correction for variation in the coronal orientation of the groove. I have attached these studies and will detail them in future articles on the operative technique.

What we have developed is a more accurate technique for referencing the trochlear groove than Whiteside’s Line. For those of you who are thinking that you are already doing the technique I’m describing because you draw in the Sulcus Line I can assure you that you are not. Sorry to be so blunt, but I’ll also go over it all in the next articles on the surgical technique. I’ll also share the technique for measuring the SL on CT and MRI scans. After that I’ll share the results of our recent studies on femoral rotational asymmetry and how it is relevant. Please “follow” me in WordPress so you don’t miss the next articles.

Published papers links:







  1. Arima J, Whiteside LA, McCarthy DS, White SE. Femoral rotational alignment, based on the anteroposterior axis, in total knee arthroplasty in a valgus knee. A technical note. J Bone Joint Surg Am. 1995;77(9):1331-4.
  2. Siston RA, Patel JJ, Goodman SB, Delp SL, Giori NJ. The variability of femoral rotational alignment in total knee arthroplasty. J Bone Joint Surg Am. 2005;87(10):2276-80. doi:10.2106/JBJS.D.02945.
  3. Middleton FR, Palmer SH. How accurate is Whiteside’s line as a reference axis in total knee arthroplasty? Knee. 2007;14(3):204-7.
  4. Victor J. Rotational alignment of the distal femur: a literature review. Orthop Traumatol Surg Res. 2009;95(5):365-72. doi:10.1016/j.otsr.2009.04.011.
  5. 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.
  6. Talbot S, Dimitriou P, Mullen M, Bartlett J. Referencing the sulcus line of the trochlear groove and removing intraoperative parallax errors improve femoral component rotation in total knee arthroplasty. Knee Surg Sports Traumatol Arthrosc. 2015:1-8. doi:10.1007/s00167-015-3668-7.
  7. Iranpour F, Merican AM, Dandachli W, Amis AA, Cobb JP. The geometry of the trochlear groove. Clin Orthop. 2010;468(3):782-8.
  8. Victor J, Van Doninck D, Labey L, Innocenti B, Parizel PM, Bellemans J. How precise can bony landmarks be determined on a CT scan of the knee? Knee. 2009;16(5):358-65.
  9. 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.
  10. Talbot S, Chao TW, Geraghty L. Combining the Sulcus Line and Posterior Condylar Axis Reduces Femoral Malrotation in Total Knee Arthroplasty. Orthopaedic Journal of Sports Medicine. 2016;4(1 suppl):2325967116S00015.


  1. Thanks for sharing.It’s very interesting finding, to say the least.I can see so much value and novel information which can definitely help to solve the rotation issue.

    Do you think there is an urgent need of redefining the rotational concept as I reckon most of the concept related to the rotation is more or less 2D based concepts?Probably at the time of developing rotational concepts, 3D measurement was not in trend. Now that we have the luxury of 3D modelling, the concept of AP axis with the history of inconsistent results does not make much sense.

    Just curious to know, Do you find any gender-based difference the curvature of SL?

    Also, I just took a glance at another interesting study of yours “Averaging rotational landmarks during total knee arthroplasty reduces component malrotation caused by femoral asymmetry.”
    As you determine the rotational pin position after averaging out the SL and PCL which seems the step ahead after validating the SL concept earlier. It would be interesting know the relationship between sizing and rotation using SL and PCL as a reference. Have you seen any difference in sizes used by taking reference of Whiteside and TEA against referencing SL and PCL??

    I will be waiting for your next blog and thanks again for writing it on Alignment which is one of my favourite topics.


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 )

Google+ photo

You are commenting using your Google+ 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