Knee pain and associated complications after intramedullary nailing of tibial shaft fracture

Tibial shaft fractures are the most common long bone fracture worldwide, with an annual incidence of 26 per 100,000 [1]. Intramedullary nailing (IMN) locked with anti-rotation screws is the first-choice treatment of displaced tibial shaft fractures in adults [2]. The method is not without complications, though. Chronic anterior knee pain at the insertion site is among the most frequently reported complications with an incidence ranging from 10% to 87% and a mean incidence of 47.4% in meta-analysis [3]. The cause of such pain remains unknown, but is believed to be multifactorial, with proposed causes being the surgical approach in relation to the patella tendon [4], placement of the incision site [5], violation of Hoffa’s fat pad [6], violation of intra-articular structures [7], nail prominence [8], nail diameter [9] and atrophy of thigh musculature [10]. Studies have also suggested that patients have a significant rate of both subjective and objective complications, where restrictions in quality of life and limitations in sports have been reported [11].

In this study, we conducted a retrospective analysis of patients who had undergone surgery with the insertion of an IMN after tibial shaft fracture. The aim of this study was to evaluate the long-term frequency of knee pain and associated complications after treatment with IMN.

METHODS

We conducted a retrospective multicentre study with the orthopaedic departments at the following five
Danish hospitals: Hvidovre Hospital, Herlev Hospital, Slagelse Hospital, Odense University Hospital and Kolding Hospital. Doctors specialised in orthopaedic surgery or attending such surgery were assigned to manage a patient-charting database search at each hospital, using the local charting database. Data including social security number, time of surgery and address were obtained. All patients who underwent surgery with reamed, locked IMN after isolated tibial shaft fracture at one of the above-mentioned orthopaedic departments between 1 November 2009 and 30 October 2014 were sought. Patients aged 18 years or older, alive and residing in Denmark at the time of follow-up were included in the study. Patient characteristics are presented in Table 1. We excluded patients who were unable to fill out the questionnaire due to concomitant physical conditions or who had undergone amputation or further surgery on the affected limb.

This was a cross-sectional cohort study where data were collected by sending out invitations and Knee Injury and Osteoarthritis Outcome Score (KOOS) questionnaires by mail to each of the patients matching the inclusion criteria. Non-responders received another letter as a reminder to the first invitation.

The study was approved by the Danish Data Protection Agency and registered with clinicaltrials.gov with registration number: NCT03649360.

The Knee Injury and Osteoarthritis Outcome Score questionnaire

We chose the KOOS questionnaire [12] for collection of patient data as it is patient-administered, the format is user friendly and it takes only about ten minutes to complete. Furthermore, it is self-explanatory and can be used as a postal survey. The questionnaire is used to assess the patient’s subjective opinion about symptoms related to the knee and other associated problems.
It is designed for patients who have experienced knee injury including meniscal injury, anterior cruciate ligament injury, tibial plateau fracture, total knee replacement and osteochondral lesions that can result in post-traumatic osteoarthritis. The questionnaire has been translated and validated in Danish and consists of five subscales: pain, symptoms, function in daily living (ADL) function in sports and recreation, and knee-
related quality of life (QoL). Each subscale has between four and 17 questions (a total of 42 questions) with each question having five response options ranging from no symptoms to severe symptoms. Each question is scored from zero to four, and a score from 0-100 is calculated: 100 indicating no symptoms and 0 indicating major symptoms.

Statistical analysis

Sample size estimation was based on an independent unequally sized two-sample t-test for the KOOS pain score between the reference population and the study group. The minimal clinically relevant difference was 6.0 points for pain, 5.0 points for symptoms, 7.0 points for ADL, 5.8 for sports and recreation and 7.0 points for QoL – set in line with the KOOS recommendations [13]. Power was set to 90% and a significance level of
p = 0.05 was chosen. The standard deviation (SD) was estimated from the pooled variance using the SD from a similar study population and the SD calculated from the KOOS confidence interval (CI) of the reference population of 533, estimating a SD of 29.3, giving a study sample size of 109. Expected dropout from non-responding was set to 40%, corresponding to a minimum of 182 patients needed.

We defined four different age groups (18-34, 35-54, 55-74 and 75-99 years) to compare age- and gender-
related differences.

Data were then collected, analysed and compared with a KOOS reference population [14]. Reference data have been published in a general population-based sample made in Southern Sweden for 840 subjects aged 18-84 years. The population was divided into four similar age groups (18-34, 35-54, 55-74, 75-99 years). This study used the same age groups for data analysis. The population did not undergo any kind of surgery before filling out the KOOS questionnaire.

Trial registration: The study was approved by the Danish Data Protection Agency. Clinical trials registration: NCT03649360.

RESULTS

A total of 391 patients were enrolled from the search. In all, 31 patients had either emigrated or had concomitant disorders that made them ineligible for study participation. A total of 24 patients had died. Invitations and questionnaires were sent out to 336 patients, and of these 113 did not return the questionnaire producing a final cohort of 223 patients (66% response rate) (Figure 1).

The mean age at the time of fracture was 47.9 years (31.6-64.2) and the mean age at follow-up was 52.1 (35.8-68.4). Time from operation to follow-up ranged from 1.7 to 6.7 years (4.2) (Table 1).

Patients in this study generally reported more serious KOOS scores than the reference population on all five subscales (Table 2). A comparison of the reference population showed that the age group of 18-34-year-olds reported the most serious difficulties. For the subgroups of pain, ADL, and especially, function in sports and recreation and QoL, they reported more difficulties when mean scores were compared with the reference group (Figure 2). When asked to indicate which degree of difficulty they had experienced while kneeling in the past week, 76.9% in the female group and 75% in the male group answered either ”severe” or ”extreme”. 66.7% in the male group reported the same degree of difficulty when asked about running and 62.5% when asked about jumping. 66.7% in the male group answered ”daily” or ”constantly” when asked how often they were made aware of their knee problem. A similar pattern was seen in the female 35-54-year-old population, where 59.3% reported the same degree of difficulties. In comparison, only 32.6% of the men in the 55-74-year group reported the same severity of symptoms.

In the 35-54-year-old population, women reported a statistically significantly poorer outcome than the
reference population in all subgroups (Figure 2). 48% answered either “totally” or “severely” when asked whether they had modified their lifestyle to avoid potentially damaging activities to the knee.

The 75-99-year-old population had 17 patients with only six in the female group, which made it insufficient for statistical analyses.

DISCUSSION

The exact cause of anterior knee pain after IMN remains unknown. The reason is thought to be multifactorial, and several proposed causes have been investigated. Whether exact causative factors arise from the injury or the operation has not been clarified. Skoog et al [15] have compared high- and low-energy injuries and reported significantly poorer outcomes for patients with tibial fracture after high-energy trauma. Katsoulis et al [3] suggested inter-operative traumas as a possible cause of long-term complications.

One cause that has been discussed in various papers is whether the transtendinous approach is more related to post-operative knee pain than the paratendinous approach. Court-Brown et al [16] documented no association between these two surgical approaches and anterior knee pain, although Keating et al [17] found a clear association between the transtendinous surgical approach and chronic anterior knee pain.

In this study, the 18-34-year-old group reported the most serious symptoms when compared to the reference population. In the subscale ”Sports and Recreation”, women and men reported a mean KOOS score of 39.29 and 40.21, respectively. In comparison, the reference population that did not undergo any surgery, reported significantly higher KOOS scores with calculated CI of 81.5-91.3 for women and 79.7-90.5 for men. 76.9% in the female group and 75% in the male group answered either ”severe” or ”extreme” when asked to indicate the degree of difficulty they had experienced while kneeling in the past week. 66.7% in the male group reported the same degree of difficulty when asked about running and 62.5% when asked about jumping. On the ”Quality of Life” subscale, the reported mean was 55.4 and 48.2 for women and men, respectively, with the confidence interval for reference population being 78.9-88.3 and 80.3-90.3, respectively. A reason for this trend might be that the younger patients are more active and therefore have higher demands for post-operative mobilisation.

In the 18-34-year-old group, 66.7% of the men reported ”daily” or ”constantly” when asked how often they were made aware of their knee problem; 32.6% of the men in the 55-74-year-old population reported the same. This may also indicate a higher level of expectations in the younger groups.

A similar pattern with symptoms being more serious in younger patients was found by Larsen et al [11]. This study also used the KOOS questionnaire to compare patients treated with IMN to a reference population with a mean of follow-up of 7.9 years. The study group in this study reported a 44% higher incidence of knee pain, a 39% higher incidence of function in daily living limitations, a 58% higher incidence of limitations in quality of life and a 60% higher incidence of limitations during sports activities. Similar to this study, the group of 18-34-year-olds reported the most severe difficulties.

Court-Brown et al [16] also reported that younger patients experienced more severe knee pain. In a retrospective study with 169 patients treated with IMN after tibia shaft fracture, 56.2% reported anterior knee pain, with 91.8% experiencing pain on kneeling and 33.7% having pain even at rest. They did not, however, find differences related to gender where this study in general reported more severe symptoms experienced by women than by men. Vaistö et al [10] also reported that women were more symptomatic than men and had a longer hospital stay after tibial nailing. One explanation for the variation in knee pain seen with gender is the prevalence of widespread pain. Widespread pain is related to age with an increase in patients over 50 years of age [18], and one study described that long-standing knee pain in women was more frequently part of a widespread pain syndrome than knee pain in men (68% versus 40%) [19]. We used the KOOS questionnaire as it has been described as a reliable and responsive tool for assessment of knee complaints [20]. One limitation associated with using this questionnaire is that it does not include data on knee disease or general health status prior to the injury, making it difficult to precisely analyse the variations seen. A possible addition to a future study may be the assessment of patient’s total body pain to separate patients with widespread pain from those with knee pain only.

The limitations of this study are its retrospective design with patients being enrolled via a database search and then conducting a cross sectional study by using the KOOS questionnaire. Whether the patients’ symptoms were more severe or the same before filling out the questionnaire was not investigated. Some of the strengths to this study are the high 66% response rate with a follow-up of up to nearly seven years, the large number of patients enrolled and, finally, the existence of a KOOS reference population.

CONCLUSIONS

Knee pain, swelling and stiffness, restrictions in quality of life and limitations in sports remain common complications after operation with the insertion of an intramedullary nail after tibia shaft fracture. When compared to a reference population, younger patients and women in general reported more difficulties. After a follow-up period of up to nearly seven years, the primary limitations were reported on ”Sports and Recreation” and ”Quality of Life”. Among the 18-34-year-olds in the ”Sport and Recreation” group, 76.9% of the women and 75% of the men indicated that they had
either experienced ”severe” or ”extreme” difficulty while kneeling in the past week.

CORRESPONDENCE: Nikolaj Erin-Madsen.
E-mail: nikolajerin-madsen@hotmail.com

ACCEPTED: 16 May 2019

CONFLICTS OF INTEREST: none. Disclosure forms provided by the authors are available with the full text of this article at Ugeskriftet.dk/dmj