Keywords
patellar tendon shortening - postoperative outcomes
Introduction
Primary total knee arthroplasty (TKA) is an extremely successful procedure for the
resolution of pain and functional impairment from end-stage knee osteoarthritis (OA).
However, complications involving the patellofemoral joint following TKA remain common
indications for revision TKA.[1] Of the described complications, patella baja is characterized by (1) distal positioning
of the patella in the femoral trochlea, (2) patellar tendon shortening (PTS), and
(3) decreased distance between the inferior pole of the patella and articular surface
of the tibia.[2] The incidence of significant PTS, defined as a decrease of 10% or greater in Insall-Salvati
ratio (ISR) after TKA, has been reported to range from 34 to 38% at 1 and 5 years
postoperatively.[3]
[4] Clinically, patients with PTS following TKA may present with anterior knee pain,
crepitus, reduced range of motion (ROM), and reduced functional outcome scores.[5]
[6] The exact etiology, however, has yet to be completely understood and is likely multifactorial.[3]
[7] Ischemic injury sustained intraoperatively via lateral retinacular release, tourniquet
use, and infrapatellar fat pad excision has been postulated to cause PTS due to peripatellar
fat pad contractures or quadriceps insufficiency.[3]
[7]
[8]
[9]
[10] Alternatively proposed etiologies of PTS include the formation of new bone, tethering
of the tendon by intra-articular fibrous bands, and scarring of the patella to the
retinaculum.[11]
[12]
In addition to the previously proposed etiologies of PTS that result in chronic shortening,
acute PTS may occur in the immediate postoperative period as a result of aberrant
soft tissue tensioning during surgical wound closure. Arthrotomy closure in flexion
has been shown to contribute to early ROM recovery and reduced physical therapy (PT)
requirements,[13]
[14] although Masri et al found no significant difference between capsular closure in
flexion versus extension.[15] Recently, Clark et al found closing either the arthrotomy and skin or skin only
in extension led to reduced postoperative patellar height when compared to in flexion;
however, patellar height reduction was not sustained at 1 year postoperatively.[16]
These findings suggest some restoration of patellar length postoperatively but fail
to provide specific details pertinent to when the greatest increase in length may
occur and over what time period.[16] Additionally, it is unclear whether any limitations in function were sustained due
to this finding or if any intervention was required to return to the patient's baseline
functional status. Understanding when the patellar tendon achieves the greatest restoration
in length in patients with PTS after TKA can help guide surgeons in determining when
conservative or more invasive interventions may be necessary. Although techniques
have been described in the literature for the management of patella baja after TKA,
there is currently no gold-standard procedure or algorithmic approach for its management.[17] The purpose of this study was to identify when postoperative patellar tendon lengthening
occurs in patients with iatrogenic patella baja secondary to significant PTS following
TKA and during what early, postoperative time period patients achieve the greatest
restoration in patellar tendon length. We hypothesize that patients will experience
significant PTS in the acute postoperative period following TKA but will subsequently
achieve patellar tendon lengthening by 6 weeks postoperatively.
Methods
Study Population
This study was evaluated by our institutional review board (IRB) and determined to
be exempt from review. A consecutive series of patients ≥18 years of age undergoing
unilateral, primary TKA for end-stage knee OA from January through April 2024 by a
single, high-volume, fellowship-trained arthroplasty surgeon were included in our
study. Patients were excluded if they (1) had a preoperative diagnoses of connective
tissue diseases, disorders associated with joint hypermobility, or diseases with pro-fibrotic
processes, (2) a history of ipsilateral knee trauma, surgery, or infection, or (3)
had preoperatively diagnosed patella baja, defined as an ISR <0.8 or patella alta,
defined as an ISR >1.2.
Data Collection
Data on patient demographics including age, sex, race, ethnicity, body mass index
(BMI), American Society of Anesthesiologist (ASA) score, age-adjusted Charlson Comorbidity
Index (CCI) score,[18] and preoperative patient-reported outcomes measures (PROMs) including the knee injury
and osteoarthritis outcome score for joint replacement (KOOS-JR) and Veteran's Rand
12 Item Health Survey (VR-12) were prospectively recorded. Active knee flexion, extension,
and total ROM, defined as extension subtracted from flexion, was measured by the operating
surgeon using a goniometer and recorded preoperatively at time of surgery sign-up
(within 1 month of TKA) and at 6-week follow-up appointments. Perioperative data including
tourniquet use and time, estimated blood loss (EBL), and tranexamic acid (TXA) use,
in addition to postoperative data including PT requirements and manipulation under
anesthesia (MUA), were also prospectively recorded.
Radiographic Measurements
Patellar tendon length was measured via ISR on preoperative, POD 0, and 6-week postoperative
radiographs as per the methods described by Insall and Salvati and validated by Cabral
et al.[19]
[20] The ISR is measured by drawing one line from the proximal tibial tubercle to the
inferior pole of the patella and dividing that distance by that measured by drawing
another line from the superior to inferior pole of the patella. Significant PTS was
defined as a decrease in the ISR by ≥10%, consistent with prior literature.[3]
[7] Measurements were performed using our institution's Picture Archiving and Communication
System (PACS) (Sectra IDS7 v.24.2).
Study Outcomes
The primary outcome of the study was the incidence of significant PTS following primary
TKA and changes in patellar tendon length from preoperative radiographs to those taken
on POD 0 and at 6 weeks postoperatively. Secondary outcomes included ROM, PT attendance,
and MUA for postoperative stiffness.
Surgical Technique
All primary TKAs were performed by one, fellowship-trained, high-volume arthroplasty
surgeon at our institution. Spinal anesthesia was utilized in all instances unless
contraindicated. After sterile preparation and draping of the patient, the operative
leg was exsanguinated prior to padded, pneumatic tourniquet inflation. A medial parapatellar
approach was utilized to gain access to the knee joint. All cuts were checked using
cutting and alignment guides prior to trialing components. Once the knee was felt
to be well aligned and stable throughout range of motion with trial components, bony
surfaces were irrigated prior to cementation of final components using pressurized
vacuum-mixed cement. All patients received P.F.C. Sigma Knee System (DePuy Synthes,
Warsaw, IN) components, and specifically, posterior stabilized femoral components.
Additionally, all patellae were resurfaced. The knee was held in 30 degrees of flexion
while the wound was closed in layers using #1 Vicryl for arthrotomy closure, 2-0 Monocryl
for subcuticular closure, and Dermabond or staples for skin closure, with a soft,
compressive dressing applied.
Postoperative Protocol
All patients were instructed to weight-bear as tolerated and met once with a physical
therapist prior to discharge. Patients were instructed to follow a detailed, home
stretching protocol with emphasis placed on active knee flexion to 90 degrees in the
first 2 weeks following surgery. Patients who did not achieve 90 degrees of active
knee flexion by 2 weeks postoperatively were prescribed formal, outpatient PT.
Statistical Analyses
Descriptive statistics were calculated first to understand the overall distribution
of the data. Independent t-tests or analysis of variance was used to compare continuous
variables for parametric data and Mann-Whitney U or Kruskal-Wallis tests for nonparametric
data. Chi-square or Fisher's exact tests were used to compare categorical data. Data
are either presented as mean (standard deviation) or count (%). Significance was determined
at P-value <0.05. All statistical analyses were done using R Studio (Version 4.1.2, Vienna,
Austria).
Final Cohort
A total of 89 patients were included in the final analysis. The mean age was 66.9 ± 8.3
years, and 61 (68.5%) patients were women. In our final cohort, 62 (69.7%) patients
identified as White and 24 (27.0%) identified as Black, while 87 (97.8%) patients
identified as not Hispanic or Latino and 2 (2.3%) identified as either Hispanic or
Latino. Mean ASA score was 2.5 ± 0.6, BMI was 33.0 ± 5.6 kg/m2, and age-adjusted CCI score was 3.9 ± 1.9. Preoperatively, the mean KOOS-JR score
for our cohort was 42.7 ± 20.8 and VR-12 score was 30.5 ± 15.5 ([Table 1]). Preoperative ISR was 1.1 ± 0.2 while active ROM was 101 ± 19.6 degrees, specifically
4.4 ± 4.5 degrees of extension and 107 ± 13.4 degrees of flexion ([Table 2]).
Table 1
Demographics of all patients who underwent primary total knee arthroplasty (n = 89)
|
All patients
|
|
(n = 89)
|
|
Age (years)
|
66.9 ± 8.3
|
|
Sex
|
|
|
Man
|
28 (31.5)
|
|
Woman
|
61 (68.5)
|
|
Race
|
|
|
Black
|
24 (27.0)
|
|
Other
|
3 (3.4)
|
|
White
|
62 (69.7)
|
|
Ethnicity
|
|
|
Hispanic or Latino
|
2 (2.3)
|
|
Not Hispanic or Latino
|
87 (97.8)
|
|
Laterality
|
|
|
Left
|
39 (43.8)
|
|
Right
|
50 (56.2)
|
|
ASA
|
2.5 ± 0.6
|
|
BMI
|
33.0 ± 5.6
|
|
CCI
|
3.9 ± 1.9
|
|
Preoperative KOOS-JR
|
42.7 ± 20.8
|
|
Preoperative VR-12
|
30.5 ± 15.5
|
Abbreviations: ASA, American Society of Anesthesiologist's physical status classification
score; BMI, body mass index; CCI, age-adjusted Charlson comorbidity index; KOOS-JR,
Knee Injury and Osteoarthritis Outcome Score for Joint Replacement; VR-12, Veterans
Rand 12 Item Health Survey.
Note: Values given as mean ± SD or N (%).
Table 2
Clinical characteristics of all patients who underwent primary total knee arthroplasty
(n = 89)
|
All patients
|
|
(n = 89)
|
|
Anesthesia
|
|
|
General
|
5 (5.6)
|
|
Spinal
|
84 (94.4)
|
|
Cut to close (min)
|
84.3 ± 12.6
|
|
EBL (mL)
|
5.3 ± 9.5
|
|
Tourniquet use
|
|
|
Yes
|
89 (100)
|
|
Time (min)
|
84.4 ± 14.9
|
|
TXA use
|
|
|
No
|
37 (41.6)
|
|
Yes
|
52 (58.4)
|
|
Dose (mg)
|
981 ± 169
|
|
Significant postoperative PTS
|
|
|
No
|
35 (39.3)
|
|
Yes
|
54 (60.7)
|
|
ISR
|
|
|
Preoperatively
|
1.1 ± 0.2
|
|
POD 0
|
1.0 ± 0.2
|
|
6 weeks PO
|
1.1 ± 0.2
|
|
∆ ISR (%)
|
|
|
Preoperatively to POD 0
|
−13.6 ± 14.1
|
|
POD 0 to 6 weeks PO
|
18.0 ± 16.7
|
|
Preoperatively to 6 weeks PO
|
0.5 ± 10.6
|
|
Flexion (degrees)
|
|
|
Preoperatively
|
107 ± 13.4
|
|
6 weeks PO
|
110 ± 19.0
|
|
Extension (degrees)
|
|
|
Preoperatively
|
4.4 ± 4.5
|
|
6 weeks PO
|
1.9 ± 4.1
|
|
Total ROM (degrees)
|
|
|
Preoperatively
|
101 ± 19.6
|
|
6 weeks PO
|
108 ± 18.8
|
|
Physical therapy required
|
|
|
No
|
61 (68.5)
|
|
Yes
|
28 (31.5)
|
|
MUA
|
|
|
No
|
88 (98.9)
|
|
Yes
|
1 (1.1)
|
Abbreviations: EBL, estimated blood loss; ISR, Insall-Salvati ratio; MUA, manipulation
under anesthesia; PO, postoperatively; POD, postoperative day; PTS, patellar tendon
shortening; ROM, range of motion; TXA, tranexamic acid.
Values given as mean ± SD or N (%).
Results
Of the 89 patients included in the final cohort, 54 (60.7%) patients comprised the
significant PTS cohort while 35 (39.3%) patients comprised the insignificant PTS cohort
([Table 3]). Patients in the significant PTS cohort were more likely to have undergone right
primary TKA; however, no significant demographic differences in patient age, sex,
race, ethnicity, BMI, ASA or CCI score, and preoperative KOOS-JR and VR-12 scores
were found to exist between cohorts.
Table 3
Demographics of all patients who underwent primary total knee arthroplasty by presence
of significant patellar tendon shortening on postoperative day 0 (n = 89)
|
Insignificant PTS
|
Significant PTS
|
P-value
|
|
(n = 35)
|
(n = 54)
|
|
Age (years)
|
67.2 ± 7.2
|
66.7 ± 8.9
|
0.970
|
|
Sex
|
|
|
0.996
|
|
Man
|
11 (31.4)
|
17 (31.5)
|
|
|
Woman
|
24 (68.6)
|
37 (68.5)
|
|
|
Race
|
|
|
0.612
|
|
Black
|
9 (25.7)
|
15 (27.8)
|
|
|
Other
|
2 (5.7)
|
1 (1.9)
|
|
|
White
|
24 (68.6)
|
38 (70.4)
|
|
|
Ethnicity
|
|
|
0.250
|
|
Hispanic or Latino
|
0 (0.0)
|
2 (3.7)
|
|
|
Not Hispanic or Latino
|
35 (100)
|
52 (96.3)
|
|
|
Laterality
|
|
|
0.041
|
|
Left
|
20 (57.1)
|
19 (35.2)
|
|
|
Right
|
15 (42.9)
|
35 (64.8)
|
|
|
ASA
|
2.4 ± 0.5
|
2.5 ± 0.6
|
0.390
|
|
BMI
|
32.7 ± 5.8
|
33.2 ± 5.5
|
0.756
|
|
CCI
|
3.9 ± 1.9
|
3.9 ± 1.8
|
0.990
|
|
Preoperative KOOS-JR
|
41.4 ± 21.0
|
43.8 ± 21.0
|
0.275
|
|
Preoperative VR-12
|
30.4 ± 16.1
|
30.5 ± 15.3
|
0.953
|
Abbreviations: ASA, American Society of Anesthesiologist's physical status classification
score; BMI, body mass index; CCI, age-adjusted Charlson comorbidity index; KOOS-JR,
Knee Injury and Osteoarthritis Outcome Score for Joint Replacement; VR-12, Veterans
Rand 12 Item Health Survey.
Note: Values given as mean ± SD or N (%).
Preoperatively, no significant differences in ISR, active flexion, extension, or total
ROM were found to exist between cohorts ([Table 4]). Perioperatively, 33 (94.3%) patients in the insignificant PTS cohort and 51 (94.4%)
patients in the significant PTS cohort received spinal anesthesia (P = 0.975). Cut to close time, EBL, tourniquet time and use, as well as TXA dosage
and use were similar between cohorts. On POD 0, ISR was significantly less for patients
in the significant PTS cohort at 0.9 ± 1.2 when compared to 1.1 ± 0.1 for patients
in the insignificant PTS cohort (P < 0.001). When compared to preoperative imaging, patients in the significant PTS
cohort experienced a decrease in their ISR by 21.9 ± 8.7% while patients in the insignificant
PTS cohort experienced a decrease of 0.8 ± 10.9% (P < 0.001) on POD 0.
Table 4
Clinical characteristics of all patients who underwent primary total knee arthroplasty
by presence of significant patellar tendon shortening on postoperative day 0 (n = 89)
|
Insignificant PTS
|
Significant PTS
|
P-value
|
|
(n = 35)
|
(n = 54)
|
|
Anesthesia
|
|
|
0.975
|
|
General
|
2 (5.7)
|
3 (5.6)
|
|
|
Spinal
|
33 (94.3)
|
51 (94.4)
|
|
|
Cut to close (min)
|
85.5 ± 11.6
|
83.4 ± 13.5
|
0.411
|
|
EBL (mL)
|
3.1 ± 7.3
|
7.1 ± 10.6
|
0.053
|
|
Tourniquet use
|
|
|
|
|
Yes
|
35 (100)
|
54 (100)
|
|
|
Time (min)
|
89.4 ± 17.0
|
81.9 ± 13.5
|
0.198
|
|
TXA use
|
|
|
0.261
|
|
No
|
12 (34.3)
|
23 (65.7)
|
|
|
Yes
|
25 (46.3)
|
29 (53.7)
|
|
|
Dose (mg)
|
960 ± 96.2
|
997 ± 210
|
0.670
|
|
ISR
|
|
|
|
|
Preoperatively
|
1.1 ± 0.2
|
1.2 ± 0.2
|
0.306
|
|
POD 0
|
1.1 ± 0.1
|
0.9 ± 0.2
|
<0.001
|
|
6 weeks PO
|
1.2 ± 0.2
|
1.1 ± 0.2
|
0.253
|
|
∆ ISR (%)
|
|
|
|
|
Preoperatively to POD 0
|
−0.8 ± 10.9
|
−21.9 ± 8.7
|
<0.001
|
|
POD 0 to 6 weeks PO
|
7.6 ± 12.0
|
25.0 ± 15.8
|
<0.001
|
|
Preoperatively to 6 weeks PO
|
5.7 ± 7.9
|
−2.9 ± 10.9
|
<0.001
|
|
Flexion (degrees)
|
|
|
|
|
Preoperatively
|
107 ± 14.2
|
106 ± 13.0
|
0.667
|
|
6 weeks PO
|
110 ± 15.0
|
110 ± 21.4
|
0.668
|
|
Extension (degrees)
|
|
|
|
|
Preoperatively
|
4.0 ± 4.4
|
4.6 ± 4.6
|
0.578
|
|
6 weeks PO
|
2.1 ± 3.7
|
1.8 ± 4.3
|
0.434
|
|
Total ROM (degrees)
|
|
|
|
|
Preoperatively
|
103 ± 16.8
|
100 ± 21.4
|
0.643
|
|
6 weeks PO
|
105 ± 24.0
|
110 ± 14.4
|
0.329
|
|
Physical therapy required
|
|
|
0.214
|
|
No
|
23 (65.7)
|
38 (70.4)
|
|
|
Yes
|
12 (34.3)
|
16 (29.6)
|
|
|
MUA
|
|
|
0.418
|
|
No
|
35 (100)
|
53 (98.2)
|
|
|
Yes
|
0 (0.0)
|
1 (1.1)
|
|
Abbreviations: EBL, estimated blood loss; ISR, Insall-Salvati ratio; MUA, manipulation
under anesthesia; PO, postoperatively; POD, postoperative day; PTS, patellar tendon
shortening; ROM, range of motion; TXA, tranexamic acid.
Note: Values given as mean ± SD or N (%).
From POD 0 to 6 weeks postoperatively, significant PTS patients experienced an increase
in their ISR by 25.0 ± 15.8% while patients in the insignificant PTS cohort experienced
an increase of 7.6 ± 12.0% (P < 0.001). When comparing change in ISR from preoperative imaging to 6 weeks postoperative
imaging, ISR decreased by 2.9 ± 10.9% for patients in the significant PTS cohort and
increased by 5.7 ± 7.9% for patients in the insignificant PTS cohort (P < 0.001) but at 6 weeks postoperatively, there was no significant difference in ISR
between cohorts. Additionally, total range of motion, flexion, or extension was similar
between groups at 6 weeks postoperatively. Similarly, 12 (34.3%) patients in the insignificant
PTS cohort participated in formal, outpatient PT while 16 (29.6%) patients in the
significant PTS cohort did (P = 0.214). One (1.1%) patient in the significant PTS cohort underwent MUA for knee
stiffness while no patients in the insignificant PTS cohort required MUA (P = 0.418).
Discussion
In the present study, 54 (60.7%) patients underwent significant PTS following primary
TKA while 35 (39.3%) patients did not. Patients with significant PTS were more likely
to experience a decrease in their ISR immediately following TKA (−21.9 ± 8.7% versus
−0.8 ± 10.9%, P < 0.001). Although by 6 weeks postoperatively, there were no significant differences
between cohorts with regard to ISR, there were significant difference in ISR dynamics
between groups; differences were seen between group from POD 0 to 6 weeks postoperatively
(25.0 ± 15.8% versus 7.6 ± 12.0%, P < 0.001), and from preoperatively to 6 weeks postoperatively (−2.9 ± 10.9% versus
5.7 ± 7.9%, P < 0.001). ROM, PT needs, and MUA rates did not differ between groups. Although ISR
changes occurred perioperatively, POD 0 PTS did not predict long-term PTS or patella
baja.
Although primary TKA remains an extremely successful procedure for the definitive
management of end-stage knee OA, complications following TKA that involve the patellofemoral
joint remain common causes of revision TKA.[1] PTS, defined as a decrease of 10% or greater in ISR and a characteristic of patella
baja, has been reported to range from 34 to 38% at 1 and 5 years postoperatively.[3]
[4] However, the etiologies of PTS following TKA have yet to be fully understood; PTS
in the acute, postoperative period may be the result of excessive tissue tensioning
during arthrotomy closure. Ischemic injury secondary to tourniquet use, infrapatellar
fat pad excision, and lateral retinacular release have also been postulated to contribute
to chronic PTS.[3]
[7]
[8]
[9]
[10] Clinically, patients with PTS following TKA may describe anterior knee pain, crepitus,
reduced ROM, and decreased PROMs, possibly necessitating prolonged PT, MUA, or ultimately
revision TKA.[5]
[6] To our knowledge, this is the first study to identify the incidence of significant
PTS in the acute, postoperative period and to what degree postoperative patellar tendon
lengthening occurs in patients during this time period.
The incidence of PTS following TKA has been reported in current literature to range
from 12 to 38%.[3]
[21]
[22] Acute PTS following TKA has been proposed to result from aberrant soft tissue tensioning
during surgical wound closure; however, Masri et al found no significant difference
to exist in early ROM recovery and PT requirements when the capsular closure was done
in flexion versus extension.[15] In our study, all TKAs were performed by the same surgeon to minimize the variability
between cases in surgical technique with all knees closed similarly, step by step.
However, 54 (60.7%) patients in our study were still found to have had significant
PTS immediately following surgery, though there were no differences by 6 weeks post-surgery.
Our findings, in combination with those of Masri et al, suggest additional etiologies
of acute PTS to exist for these patients besides surgical closure technique. Alternatively,
Noyes et al proposed decreased strength of the quadriceps femoris to cause contracture
of the patellar tendon.[7] Our study did not include quadriceps strength as a direct outcome measure; however,
active knee extension at 6 weeks postoperatively may serve as an indirect measurement
for quadriceps strength since patients in both cohorts were able to extend their knees
greater than preoperatively.
Although the exact cause of acute PTS following TKA remains unclear, no significant
differences in 6-week ISR, active ROM, or PT requirements were observed between patients
with and without early PTS. These findings suggest that patients with significant
PTS were able to regain patellar tendon length without additional intervention. As
suggested by Davies et al, preoperative pain and degeneration secondary to OA may
lead to shortening of the tendon with correction occurring after surgery as function
improves.[4] Our standard postoperative protocol for primary TKA patients involves a series of
home stretching exercises that patients are taught prior to surgery and instructed
to perform daily upon discharge; patients are not prescribed PT unless they are stiff
at their 2-week postoperative clinic visit. Our protocol may not only benefit patients'
patellar tendon length by limiting the extent to which their quadriceps decondition
but may also help patients to regain their ROM and promote movement, reducing the
acute inflammatory processes that may contribute to PTS following TKA.[11]
[12]
Although this is the first prospective study to determine the incidence of significant
PTS following primary TKA in the acute postoperative period and identify when patellar
tendon lengthening occurs in these patients, it is not without several potential limitations.
This was a prospective, cohort study performed at a single institution by one arthroplasty
surgeon. Therefore, the generalizability of our findings may be limited to patients
who undergo primary TKA using similar surgical techniques and do not allow for comparisons
to be made between alternative approaches and techniques. Additionally, the overall
sample size of the present study was relatively small. However, the power analysis
conducted prior to the start of our study indicated that 89 patients were adequate
to power our study and appropriately identify differences between cohorts. It is also
possible that no interventions such as manipulation under anesthesia (MUA) were required,
thus skewing our findings. Lastly, despite having a minimum of 6-week clinical and
radiographic follow-up for all patients, our follow-up time is too short to characterize
the potential long-term impact of acute PTS on PROMs.
Conclusion
Based on the findings of our study, majority of patients undergoing primary TKA will
experience significant PTS in the acute, postoperative period. However, we found that
by 6 weeks postoperatively, these patients will appreciate patellar tendon lengthening
and resolution of acute PTS without requiring additional PT or experiencing deficits
in ROM.
