Sage Journals: Discover world-class research

Abstract

Background:

Studies on the coronal plane alignment of the knee (CPAK) classification after high tibial osteotomy (HTO) are scarce.

Purpose:

To (1) clarify the distribution of the CPAK phenotype in patients who underwent HTO and (2) compare the clinical results according to the CPAK phenotypes.

Study Design:

Case Series; Level of evidence, 4.

Methods:

Overall, 171 knees of patients with HTO were retrospectively evaluated. The primary outcome was to investigate the serial changes in the CPAK preoperatively and 2 years postoperatively. The secondary outcome was to assess the subcategories of the knee injury and osteoarthritis outcome score (KOOS) and postoperative change of KOOS according to the CPAK.

Results:

The most common pre- and postoperative CPAK phenotype types were I (62%), II (22.8%), IV (8.8%), VI (36.7%), V (18.9%), and IX (13%). The distribution of preoperative CPAK significantly changed after HTO. KOOS subscales showed no significant differences according to the CPAK phenotypes.

Conclusion:

The most common pre- and postoperative CPAK phenotypes were I and VI. No significant differences in the KOOS subscales were observed according to CPAK phenotypes following HTO.

Keywords

coronal plane alignment of the knee classification high tibial osteotomy

The coronal plane alignment of the knee (CPAK) classification is a simple tool for assessing lower extremity alignment, as the CPAK phenotype is determined by measuring only the medial proximal tibial angle (MPTA) and the lateral distal femoral angle (LDFA).¹⁰ Since initially reported by McDessi, several registries have been described in other countries.^{4,8,11,12,15,19,24,28}

The CPAK classification has been mentioned in the literature after total knee arthroplasty (TKA) or unicompartmental knee arthroplasty (UKA).^1,3,7,26 Changes in the arithmetic hip-knee-ankle (aHKA) and joint line obliquity (JLO) in CPAK classification before and after TKA results are associated with poor postoperative outcomes.⁷ In robot-assisted TKA for CPAK type I osteoarthritis, kinematic alignment achieved knee balance in extension without soft tissue release in a higher percentage of patients than mechanical alignment.¹ Therefore, the alignment changes by TKA or UKA contribute to poor results. In contrast, alignment changes after high tibial osteotomy (HTO), a realignment surgery, contribute to good results. Although few reports have described CPAK classification after HTO,^2,18 only radiological evaluations have been reported, with no reports on the association between clinical outcomes.

This study aimed to (1) clarify the distribution of CPAK phenotypes in patients who underwent HTO and (2) compare the clinical results according to the CPAK phenotypes.

We hypothesized that (1) the postoperative distribution of the CPAK phenotype would be significantly different from the preoperative status, and (2) the neutral alignment (CPAK type V) would yield better clinical results than the other CPAK phenotypes.

Methods

The present study was approved by our institution's ethics committee (E22-0166). Standard informed consent was obtained from all patients participating in the present study.

This was a retrospective study of patients who underwent HTO between November 2011 and December 2022. This study comprised patients who completed at least 2 years of postoperative follow-up without sequential loss of clinical or radiological evaluations. The inclusion criteria were as follows: (1) medial compartment osteoarthritis; Kellgren-Lawrence grades 2 to 4; (2) concurrent varus deformity of the lower limb; weightbearing line ratio (WBLR) <50%; and (3) cases with an acceptable alignment range at 3 months postoperatively. The exclusion criteria were as follows: (1) secondary osteoarthritis; (2) association with ligament reconstruction; (3) infection; and (4) revision of HTO.

Clinical evaluations were performed using the knee injury and osteoarthritis outcome score (KOOS)²² preoperatively and at 2 years postoperatively. Postoperative changes in the KOOS were evaluated to detect true change.

Radiological Evaluation

Long-leg standing radiographs were obtained for all participants. The radiological parameters were measured in Digital Imaging and Communications in Medicine images using validated software (SurgiMap). Measurements of radiological parameters were taken by 2 observers (M.K. and K.Y.). Two observers evaluated 20 cases at 2-week intervals. Paley and Pfeil²⁰ performed the following measurements: the angle between the mechanical axis of the femur and the LDFA; the angle between the mechanical axis of the tibia and the MPTA (Figure 1).⁸ Subsequently, aHKA and JLO were calculated arithmetically (Figure 2).¹⁰ The aHKA is calculated as MPTA _ LDFA. Neutral alignment is defined as an aHKA between +2° and −2°, varus alignment as aHKA <–2°, and valgus alignment as aHKA >+2°. JLO is calculated as MPTA + LDFA, and the joint line is labeled parallel or neutral when equal to 180°. The orientation of the joint line is defined as neutral when the JLO is between 177° and 183°, apex distal when the JLO is <177°, and apex proximal when the JLO is >183°. The CPAK classification comprises 9 groups, determined by aHKA and JLO values (Figure 2).

Figures 1.

Coronal plane alignment of the knee (CPAK) classification.

Figure 2.

LDFA (lateral distal femoral angle) and MPTA (medial proximal tibial angle) LDFA is defined as the angle between the mechanical axis of the femur and the distal femoral joint line. MPTA is defined as the angle between the mechanical axis of the tibia and the proximal tibial joint line.

Surgical Technique and Postoperative Rehabilitation

All operations were performed or supervised by a single surgeon (M.K.). Before HTO, diagnostic arthroscopy was performed to verify the correct indication. Arthroscopic synovectomy was performed when necessary, and arthroscopic lateral release was done in all cases. Partial meniscectomy and medial meniscus centralization were performed for degenerative tears in the medial meniscus and in case of medial meniscus extrusion, respectively.⁶ Abrasion chondroplasty or multiple drilling was performed using a 2.0-diameter Kirschner wire for chondral defects in the medial compartment of the knee. In contrast, autologous chondrocyte implantation was not done in this study.

Based on recommendations from the AO Foundation Expert group,^21,25 biplanar osteotomy was performed in all patients. The osteotomy was stabilized using a long-type locking plate (TomoFix, Synthes; or TriS, Olympus Terumo Biomaterials). An artificial bone graft (OSferion 60; Olympus Terumo Biomaterials) was inserted into the osteotomy gap to start early weightbearing.²⁷ The usual target point of correction was approximately 62.5% of the WBLR, named the Fujisawa point.⁵ The acceptable range of postoperative coronal alignment was defined as neutral or valgus for aHKA≥–2, based on the boundaries of the CPAK classification (Figure 2). Postoperative rehabilitation began the day after surgery, with active and passive range of motion exercises and muscle-strengthening exercises. Range of motion training was continued until a maximum flexion angle of at least 130° was obtained within 3 weeks of surgery. Standing exercises were also initiated as soon as possible. Notably, 1 week after surgery, the patients were allowed to begin half-weightbearing exercises with walker equipment, and full-weightbearing walking was allowed 2 weeks after surgery. Return to normal sports activities, including jogging, was allowed after verification of bone union at the osteotomy site. The strategy has been previously reported in the literature.⁹

Statistical Analysis

The sample size was calculated using G-power 3.1.9.6 (Dυsseldorf, Germany). Based on the a priori power analysis, a minimum of 162 patients were required (α = .05; β = 0.95).

CPAK phenotype rates before and after HTO were compared using the McNemar test. Furthermore, this study investigated changes in CPAK phenotype pre- and postoperatively.

Pre- and postoperative KOOS subscales were compared across CPAK phenotypes using an analysis of variance. The intraclass correlation coefficient (ICC) was used to assess inter- and intraobserver agreement for LDFA and MPTA angle measurements. The 2-way random model was used to calculate ICCs with 95% CIs. The ICC was interpreted as follows: <0.50, poor; between 0.50 and 0.75, moderate; between 0.75 and 0.90, good; >0.90, excellent. P < .05 was considered statistically significant. The data were analyzed using SPSS Version 26.0 software (IBM Corp). Descriptive statistics were expressed as means and standard deviations for numerical variables and as numbers and percentages for categorical variables.

Results

Overall, 352 knees underwent HTO; however, some cases were excluded due to a lack of preoperative assessment (n = 1), secondary osteoarthritis (n = 5), or ligament reconstruction (n = 2) (Figure 3). Therefore, 344 knees were selected. Cases of infection (n = 3), revision of HTO (n = 3), undercorrection at postoperative 3 months (n = 13), loss of follow-up over 2 years (n = 154) were further excluded. Finally, 171 knees were included in the study (Figure 3). Among them, 68 were men, and 103 were women; the mean age was 61 ± 9.7 years, and the mean body mass index (BMI) was 26.2 ± 4 kg/m² (Table 1). There were no significant differences between the included patients and those lost to follow-up in descriptive data (mean age: 61 vs 58 years; height: 160.3 vs 163.1 cm; weight: 67.8 vs 68.8 kg; BMI: 26.2 vs 25.8 kg/m²). There were 6 knees (3.5%) in grade 1, 66 knees (38.6%) in grade 2, 75 knees (43.9%) in grade 3, and 24 knees (14%) in grade 4, based on the Kellgren-Lawrence (KL) classification. All patients preferred osteotomy to arthroplasty because they wanted to return to sports.

Figure 3.

Flowchart of patient selection and inclusion.

Table 1

Descriptive Data ^a

Parameter
Age, years		61 ± 9.7 (37 to 83)
Sex, male/female		68/103
Laterality, right/left		69/106
Height, cm		160.2 ± 8.2 (142 to 181)
Weight, kg		67.8 ± 13.7 (34 to 105)
BMI, (Kg/m²		26.2 ± 4 (17 to 40)
KL grade	1	6 (3.5)
	2	66 (38.6)
	3	75 (43.9)
	4	24 (14)
MPTA, deg		84.2 ± 3 (70 to 90)
LDFA, deg		88.3 ± 2.1 (84 to 98)
aHKA, deg		–4.1 ± 3.3 (–18 to 2)
JLO, deg		172.5 ± 4 (158 to 184)

Data are presented as mean ± SD (range) or n (%). aHKA, arithmetic hip-knee-ankle angle; BMI, body mass index; JLO, joint line obliquity; KL, Kellgren-Lawrence grade; LFDA, lateral distal femoral angle; MPTA, medial proximal tibial angle.

The mean preoperative MPTA and LDFA were 84.2°± 3° and 88.3°± 2.1°, respectively (Table 1). The mean pre- and postoperative aHKA and JLO were −4.1°± 3.3° and 172.5°± 4°, respectively. The inter- and intraobserver reliability for the measurement of radiographic parameters was satisfactory, and the mean values of LDFA and MPTA were 0.963 versus 0.950 (95% CI, 0.920 to 0.980 vs 0.894 to 0.979) and 0.968 versus 0.945 (95% CI, 0.83 to 0.92 vs 0.883 to 0.977), respectively.

The most common pre- and postoperative CPAK phenotypes were I (62%) and VI (36.7%) (Table 2). The preoperative CPAK phenotype changed mainly to CPAK type VI at 2 years postoperatively, as the postoperative MPTA increased after HTO. Significant differences were observed in the distribution of the pre- and postoperative CPAK phenotypes (P < .001). Pre- and postoperative changes of CPAK phenotypes are shown in Table 3. Almost all preoperative CPAK phenotypes changed to CPAK VI after HTO. However, CPAK IV did not change to CPAK VI; instead, it mainly changed to CPAK IX after HTO.

Table 2

Comparison Between Pre- and Postoperative CPAK Phenotypes ^a

CPAK type	Preop, n = 171	Postop 2 Years, n = 169	P
I	106 (62)	4 (2.4)	.000 ^c
II	39 (22.8)	19 (11.2)	.005^c
III	1 (0.6)	22 (13)	<.001 ^b
IV	15 (8.8)	3 (1.8)	.012 ^b
V	9 (5.3)	32 (18.9)	<.001 ^c
VI	0 (0)	62 (36.7)	.000 ^c
VII	1 (0.6)	0 (0)	<.001 ^c
VIII	0 (0)	5 (3)	.000 ^c
IX	0 (0)	22 (13)	.000 ^c

Data are presented as n (%). CPAK, coronal plane alignment of knee classification; Postop, postoperative; Preop, preoperative.

P < .05.

P < .01 by McNemar test.

Table 3

Comparison of CPAK Phenotypes Between Pre- and Postoperative 2 Years ^a

		Postop 2 Years
	CPAK, Type	Type I	Type II	Type III	Type IV	Type V	Type VI	Type VIII	Type IX	Total
Preop	I	5 (4.6)	15 (13.8)	13 (13.9)	5 (4.6)	19 (17.9)	38 (34.9)	2 (1.8)	9 (8.3)	106
	II	0 (0)	4 (10.5)	7 (18.4)	0 (0)	6 (15.8)	20 (51.2)	0 (0)	2 (5.3)	39
	III	0 (0)	0 (0)	1 (100)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	1
	IV	0 (0)	0 (0)	0 (0)	0 (0)	3 (20)	2 (13.3)	1 (6.7)	9 (60)	15
	V	0 (0)	0 (0)	1 (11.1)	0 (0)	1 (11.1)	3 (33.3)	2 (22.2)	2 (22.2)	9
	VII	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	1 (100)	0 (0)	0 (0)	1
	Total	5	19	22	5	29	63	5	22	171

Data are presented as n (%). CPAK, coronal plane alignment of knee classification; Postop, postoperative; Preop, preoperative.

The pre- and postoperative clinical results, stratified by CPAK phenotype, are shown in Figure 4. No significant differences were observed in the pre- or postoperative KOOS subscales among the CPAK phenotypes. However, the postoperative KOOS subscale of varus alignment (CPAK I and IV) was lower than that of the other CPAK phenotypes.

Figure 4.

Comparison of pre- and postoperative clinical results according to the CPAK phenotype. No significant differences were observed in the preoperative or postoperative KOOS subscales among the CPAK phenotypes.

Discussion

The main findings of this study included the following: (1) the postoperative distribution of the CPAK phenotype was significantly different from the preoperative status, and (2) the clinical results of the neutral alignment (CPAK V) were not significantly better than those of the other CPAK phenotypes. Therefore, the first hypothesis was confirmed, whereas the second hypothesis was rejected.

The most common types were I (62%), II (22.8%), IV (8.8%), and VI (36.7%) for preoperative CPAK and V (18.9%) and IX (13%) for postoperative CPAK. Almost all preoperative CPAK phenotypes changed to CPAK VI after HTO. However, CPAK IV did not change to CPAK VI; instead, it mainly changed to CPAK IX after HTO. The postoperative change to CPAK IX was due to the large LDFA. CPAK IV was constructed by a large LDFA (>90°) and a small MPTA (<85°). There were severe varus alignment cases in CPAK IX. Therefore, a large correction of the MPTA is required in HTO. JLO became apex proximal (JLO >183°) because JLO was calculated by the sum of MPTA and LDFA. As a result, preoperative CPAK IV tended to progress to CPAK IX after HTO. Ryu et al. recently reported similar findings.²³ Cases with CPAK IX should undergo double-level osteotomy to maintain neutral JLO.¹⁶

There are several registries of CPAK classification worldwide, and the characteristics of lower limb alignment vary between countries. However, only a few reports of CPAK classification after HTO are available in the literature. Mitterer et al¹⁴ reported that postoperative CPAK classifications shifted significantly across all osteotomy types, with minimal retention of preoperative types. Although most procedures achieved correction within the target HKA range, neutral JLO restoration was observed in only half of the cases, underscoring the importance of phenotype-specific planning and the potential limitations of CPAK classification.¹⁴ The restoration rate of a neutral JLO was 56.7% (97/171) in this study. These results are consistent with our findings. Maione et al¹³ reported that the surgical correction of a medial closing-wedge distal femoral osteotomy achieved a neutral mechanical axis (CPAK V and VIII), with no clinical differences across different JLO values. The postoperative KOOS subscale of varus alignment (CPAK I and IV) was lower than that of the other CPAK phenotypes (Figure 4). However, no significant differences were detected due to the few cases of varus alignment (10/171; 5.8%) after HTO in this study (see Table 2).

This study has several limitations. First, the participants' average age (61 years) was higher than that reported in previous literature.¹⁷ Second, there was a higher rate of advanced osteoarthritis (6 knees [3.5%] in grade 1, 66 knees [38.6%] in grade 2, 75 knees [43.9%] in grade 3, and 24 knees [14%] in grade 4, according to the KL classification). Clinical changes might be affected by age and osteoarthritic severity. Third, the dropout rate was approximately 50% (181/352). Finally, the CPAK classification is not assessed in the sagittal plane or for rotational alignment. A 3-dimensional deformation analysis should be performed to accurately assess lower extremity alignment.

Conclusion

The most common pre- and postoperative CPAK phenotypes were I and VI. No significant differences in the KOOS subscales were observed according to CPAK phenotypes after HTO.

Footnotes

Acknowledgements

The authors thank Shuko Nojiri for the statistical analysis and Ryoko Tarukawa for data management.

Final revision submitted December 20, 2025; accepted December 26, 2025.

The authors declared that they have no conflicts of interest in the authorship and publication of this contribution. AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto. This research was approved by the Institute Ethics Committee (E22-0166).

ORCID iD

Mitsuaki Kubota

References

Arai

Toyooka

Masuda

Kawano

Nakagawa

Kinematic alignment achieves a more balanced total knee arthroplasty than mechanical alignment among CPAK type I patients: a simulation study. J Clin Med. 2024;13(12):3596.

Cho

Nam

Park

JPY

Lee

YS.

Constitutional and postoperative joint line obliquity can predict serial alignment change after opening-wedge high tibial osteotomy: analysis using coronal plane alignment of the knee classification. Knee Surg Sports Traumatol Arthrosc. 2023;31(12):5652-5662.

Corban

van de Graaf

Chen

, et al. How often do we alter constitutional limb alignment, joint line obliquity, and Coronal Plane Alignment of the Knee (CPAK) phenotype when performing mechanically aligned TKA? Bone Jt Open. 2024;5(2):109-116.

Dragosloveanu

Capitanu

Josanu

, et al. Radiological assessment of coronal plane alignment of the knee phenotypes in the Romanian population. J Clin Med. 2024;13(14):4223.

Fujisawa

Masuhara

Shiomi

The effect of high tibial osteotomy on osteoarthritis of the knee. An arthroscopic study of 54 knee joints. Orthop Clin North Am. 1979;10(3):585-608.

Koga

Nakamura

Nakagawa

, et al. Arthroscopic centralization using knotless anchors for extruded medial meniscus. Arthrosc Tech. 2021;10(3):e639-e645.

Konishi

Hamai

Tsushima

, et al. Pre- and postoperative Coronal Plane Alignment of the Knee classification and its impact on clinical outcomes in total knee arthroplasty. Bone Joint J. 2024;106-b(10):1059-1066.

Kubota

Kaneko

Yoshida

, et al. The coronal plane alignment of the knee classification of a Japanese population indicates more varus deformity tendency compared with European countries. Knee. 2025;53:246-254.

Kubota

Kim

Kaneko

Yoshida

Ishijima

Poor accuracy of intraoperation medial proximal tibial angle measurement compared to alignment rod methods in open-wedge high tibial osteotomy for medial knee osteoarthritis. J Knee Surg. 2023;36(7):767-772.

10.

MacDessi

Griffiths-Jones

Harris

Bellemans

Chen

DB.

The arithmetic HKA (aHKA) predicts the constitutional alignment of the arthritic knee compared to the normal contralateral knee: a matched-pairs radiographic study. Bone Jt Open. 2020;1(7):339-345.

11.

MacDessi

Griffiths-Jones

Harris

Bellemans

Chen

DB.

Coronal Plane Alignment of the Knee (CPAK) classification. Bone Joint J. 2021;103-B(2):329-337.

12.

MacDessi

van de Graaf

Wood

, et al. Not all knees are the same. Bone Joint J. 2024;106-b(6):525-531.

13.

Maione

Pozzi

Menon

, et al. Preoperative CPAK phenotype does not affect clinical and radiological outcomes after medial closing-wedge distal femoral osteotomy in valgus knees at 8-year follow-up. Knee Surg Sports Traumatol Arthrosc. 2025;33(10):3686-3696.

14.

Mitterer

Huber

Pallamar

, et al. Changes of knee phenotypes following osteotomy around the knee in patients with valgus or varus deformities—a retrospective cross-sectional study. J Clin Med. 2025;14(13):4684.

15.

Mulpur

Desai

Mahajan

, et al. Radiological evaluation of the phenotype of Indian osteoarthritic knees based on the Coronal Plane Alignment of the Knee Classification (CPAK). Indian J Orthop. 2022;56(12):2066-2076.

16.

Nakayama

Schröter

Yamamoto

, et al. Large correction in opening wedge high tibial osteotomy with resultant joint-line obliquity induces excessive shear stress on the articular cartilage. Knee Surg Sports Traumatol Arthrosc. 2018;26(6):1873-1878.

17.

Ollivier

Kley

, et al. A significant rate of tibial overcorrection with an increased JLO occurred after isolated high tibial osteotomy without considering international consensus. Knee Surg Sports Traumatol Arthrosc. 2023;31(11):4927-4934.

18.

Ollivier

Gulagaci

Sahbat

Kim

Ehlinger

Should I optimize joint line obliquity in my knee procedure?

Orthop Traumatol Surg Res. 2024;110(4):103882.

19.

Pagan

Karasavvidis

Lebrun

, et al. Geographic variation in knee phenotypes based on the Coronal Plane Alignment of the Knee classification: a systematic review. J Arthroplasty. 2023;38(9):1892-1899.e1891.

20.

Paley

Pfeil

[Principles of deformity correction around the knee]. Orthopade. 2000;29(1):18-38.

21.

Pearle

Goleski

Musahl

Kendoff

Reliability of image-free navigation to monitor lower-limb alignment. J Bone Joint Surg Am. 2009;91(suppl 1):90-94.

22.

Roos

Lohmander

Ekdahl

Beynnon

BD.

Knee Injury and Osteoarthritis Outcome Score (KOOS)—development of a self-administered outcome measure. J Orthop Sports Phys Ther. 1998;28(2):88-96.

23.

Ryu

Lee

Kim

Song

Kim

HY.

Radiographic risk factors for excessive joint line obliquity after knee osteotomy for medial osteoarthritis: a phenotype-based approach. Orthop J Sports Med. 2025;13(9):23259671251351335.

24.

Senel

Eren

Sert

, et al. Phenotyping of the Turkish population according to Coronal Plane Alignment of the Knee classification: a retrospective cross-sectional study. Jt Dis Relat Surg. 2024;35(1):194-201.

25.

Takeuchi

Ishikawa

Aratake

, et al. Medial opening wedge high tibial osteotomy with early full weight bearing. Arthroscopy. 2009;25(1):46-53.

26.

Ten Noever de Brauw

Bayoumi

Ruderman

, et al. Knees with anteromedial osteoarthritis show a substantial phenotypic variation prior and following medial unicompartmental knee arthroplasty. Knee Surg Sports Traumatol Arthrosc. 2023;31(12):5579-5590.

27.

Tsukeoka

Lee

TH.

Sagittal flexion of the femoral component affects flexion gap and sizing in total knee arthroplasty. J Arthroplasty. 2012;27(6):1094-1099.

28.

Yang

Yoon

Kim

Seon

JK.

Radiologic assessment of knee phenotypes based on the Coronal Plane Alignment of the Knee classification in a Korean population. Clin Orthop Surg. 2024;16(3):422-429.

Effect of Coronal Plane Knee Alignment Classification on Clinical Outcomes After High Tibial Osteotomy

Abstract

Background:

Purpose:

Study Design:

Methods:

Results:

Conclusion:

Keywords

Methods

Radiological Evaluation

Surgical Technique and Postoperative Rehabilitation

Statistical Analysis

Results

Discussion

Conclusion

Footnotes

Acknowledgements

ORCID iD

References