Infection after knee replacement: a qualitative study of impact of periprosthetic knee infection

In the USA and UK, approximately 340,000 primary knee replacements are performed each year [1, 2]. In 2015, 98,591 primary knee replacement procedures were performed in England and Wales alone [2]. Primary knee replacement is most commonly performed for osteoarthritis [3]. Although knee replacement often improves function and decreases pain, complications after knee replacement can include long-term pain [4], periprosthetic fracture [5] and prosthetic joint infection [6]. For some complications, revision surgery is required. Periprosthetic infection occurs when the tissues surrounding the prosthesis become infected, and it accounts for 25.2% of revision procedures after knee replacement [7] with reported incidence in the UK and USA ranging from 0.5 to 2% [6, 8, 9]. In 2015 in the UK, 6104 knee revision procedures were performed, of which nearly one quarter (1420 (23%)) were for infection [2].

Periprosthetic infection is commonly treated with either one-stage or two-stage revision. One-stage revision treatment consists of one major surgery, during which the prosthesis and infected tissues are removed before a new prosthesis is re-implanted [10]. Two-stage revision treatment consist of two separate operations where the reinsertion of a new implant is delayed allowing for additional antibiotic therapy between surgeries, sometimes via an antibiotic-impregnated cement spacer [10]. Two-stage surgical revision is the most common treatment for infection after joint replacement [11], however infection is costly to health services [12] – a revision for infection is more than three times that of an aseptic revision [13] – and two-stage revisions are more expensive than one-stage revisions with the patient undergoing two major surgical procedures [14, 15].

Previous research on infection after joint replacement highlights how infection and treatment can have profoundly negative impact on all aspects of patients’ lives [16‐18]. Due to the heavy physical and psychological burden of treatment that periprosthetic infection imposes on patients and their families, there is a need for increased psychological and rehabilitative support during treatment and long-term recovery [18].

So far, research has not drawn on social science theories to extend and deepen understanding of the impact of periprosthetic infection on patients. In understanding the impact of major health and treatment events, insights from the social science of health and illness are useful, particularly regarding disruption to people’s lives, and how people negotiate and manage major change brought about by illness. Bury’s work on ‘biographical disruption’ is widely recognised as making a major contribution to the greater understanding of experiences of health and illness events. It has been used in health research to highlight how illness disrupts people’s expected life course in multiple ways, changing the structures of daily life, challenging and redefining people’s sense of identity and constructed biography [19‐22]. Bury’s work has been applied in diverse health contexts, including rheumatoid arthritis [19], long-term knee pain [23], osteoarthritis [24], multiple sclerosis, cerebral palsy, blindness [25]; Meniere’s disease [26] and cancer [27, 28]. Bury describes three aspects of disruption that take place in the experience of chronic illness: onset and the problem of recognition; emerging disability and the problem of uncertainty, and chronic illness and the mobilisation of resources [19]. In exploring the biographical impact of illness, Bury [29] conceptualises the symptoms of chronic illness as having two distinct meanings: meaning as consequence: the problems experienced by people as a result of activity restriction, social disadvantage and their impact on daily life; and meaning as significance: the connotations that illnesses carry, in a cultural context. Critique of Bury’s work often focusses on these two distinct meanings and the degrees to which each is true, with the concept of ‘normal illness’ [20] set against that which is disruptive. The concept of biographical disruption is “predicated in large part on an adult-centred model of illness denoting the shift from a normal state of health to one of illness” [25]. Based on our previous work [18], periprosthetic infection may be an example of this shifting circumstance.

In this qualitative interview study, we explore the experience of periprosthetic knee infection and its impact on patients’ lives and draw on the concept of biographical disruption to sensitise us to the meanings of significance and consequences of this major health event for individuals in the context of their social worlds.

The sample consisted of 9 patients who received one-stage revision treatment, and 7 patients who received two-stage revision treatment; 9 men and 7 women, aged 59–80 years (mean age 72 years) (Table 1).

For 14 patients, this was their first revision surgery. One had received a previous two-stage revision. With the exception of one participant whose primary knee replacement was the result of a fracture, participants’ primary replacements had been elective procedures to relieve pain associated with osteoarthritis.

The results highlighted participants’ varied experiences, in particular regarding infection onset and subsequent treatment. Infection onset occurred over periods ranging from immediately to 19 years after their primary surgery. All participants felt that infection was life-changing, describing it as “devastating”, “traumatic” and causing a “considerably restricted life”. Patients’ narratives suggest that the experience of infection can be understood as a temporal situation, which can be mapped onto Bury’s three aspects of disruption: Onset and the problem of recognition; emerging disability and the problem of uncertainty; and chronic illness and the mobilisation of resources. Participants discussed the ways in which infection and its treatment impacted on, and disrupted the life course within each aspect.

Our analysis suggests that the experience of infection and its treatment causes patients to negotiate major change brought about by this illness event, and to redefine their biographies. Participants described the disruption of taken-for-granted daily routines, uncertainty about the impact and course of infection and treatment, and anxiety and fear about the future. Illustrative quotations are referred to throughout the text and are presented in Tables 2, 3 and 4.

People diagnosed with periprosthetic infection faced an altered and unexpected situation, in which they experienced profound disruption to their life course according to Bury’s theory [19]. Infection and treatment appeared to derail people’s sense of a planned or anticipated biographical trajectory, which they found distressing. Their distress was particularly related to difficulties in getting a diagnosis of infection; a relatively sudden lack of mobility from onset of infection and throughout treatment; an uncertainty about their future; and a forced withdrawal from their social worlds. Bury’s theory provides a useful framework through which to view the narratives of patients, through each of the three aspects of disruption that take place in patients’ experience of chronic illness.

Participants made sense of their experiences of infection through the narratives that they shared, giving meaning to the events of infection that disrupted the structures of their everyday lives, and the foundations that underpin these structures [19]. Narratives often began with an uncertain, difficult or traumatic diagnosis during which some patients felt that signs of infection were “not taken seriously enough” and should have been acted upon earlier. This was followed by a disruption of taken-for-granted behaviours, and persistent uncertainty during their treatment and rehabilitation, for which they were largely unprepared. Not all participants could recall discussing the risks of infection with their doctor, or receiving information about infection, before their primary joint replacement. While this apparent lack of information may have enabled patients to remain calm and positive about their primary joint replacement, it may also have increased the likelihood of patients’ lack of preparedness when infection was later diagnosed. However, it is also possible that patients did not expect to be one of the 1% of primary knee replacement patients who develop an infection.

Periprosthetic infection can be life-changing, both physically and psychologically, and placing pressure on patients’ social and supportive relationships. The impact on participants’ mobility and physical function restricted their ability to participate in social roles and events as they had previously, subsequently affecting their sense of personal identity. Some participants were forced to adapt their home environments or move home entirely to cope with their recovery. Participants described disruption to their personal identity and relationships with others, in a context where infection and treatment precluded their former positive experiences and meanings [19, 22]. Withdrawal from social relationships as well as increasing social isolation are important features of chronic illness [37]. Increasing social isolation and dependency on their families and wider social networks, led participants to describe how they felt they were being a burden to those who cared for them. Throughout participants’ narratives uncertainty is evident - at diagnosis, treatment, rehabilitation and in concerns about eradication or return of infection. In their narratives, participants incorporated this difficult and traumatic illness event into their biography, but they often remained uncertain regarding their future. The patients in this study described a persistent need for support throughout diagnosis, treatment and recovery. Despite the relatively small number of cases of periprosthetic infection each year, the impact on patients’ lives is disproportionately adverse.

Despite its widespread application, biographical disruption has been debated [24, 25]. Yet notwithstanding these alternate views, it remains a powerful sensitising concept to the meanings of significance and consequences of periprosthetic infection for individuals in the context of their social worlds. In understanding the significance and consequence of periprosthetic infection, we suggest that its onset and treatment represents a completely unexpected and anomalous event which profoundly disrupts the lifecourse. This is the first study to explore the impact of one-stage and two-stage revision treatment for periprosthetic knee infection. Previous qualitative work has explored periprosthetic hip infection [18] and surgical site infection [38]. Our study also indicates that infection impacts on all aspects of patients’ lives [16‐18], but also explored the relationship between the experience of infection and treatment, and biographical disruption. Bury’s work on biographical disruption focuses on the onset and the perception of illness [19]: our work extends this by also focusing on the treatment of, and recovery from illness, which in the context of prosthetic joint infection arguably has the greatest impact on patients. While we acknowledge that the onset of periprosthetic infection is distressing and impactful for patients, they remain largely unprepared for the potentially lengthy and complex treatment process and recovery period. This impacts on patients’ ability to mobilise resources in order to cope with their treatment and recovery, and the participants in this study described feeling irrational, anxious and in low mood, and very aware of the burden they placed on their supportive networks. Our findings draw parallels with work describing how feelings of hopelessness are reported among patients with surgical site infections [38]. Both the loss of identity and independence established in this study are similar to those found in other chronic illnesses [39, 40]. The unpredictability and uncertainty of diagnosis and treatment of periprosthetic infection draw parallels with patients’ perceptions of multiple sclerosis [41], HIV [42‐44], and cancer [45]. The disruption to the life course that patients in this study experienced is similar to those experiencing HIV and cancer diagnoses [46‐48]. Uncertainty is widely recognised in qualitative studies of illness experience literature, and patients’ uncertainty is greatest during the diagnosis phase and when outcomes are unknown or unpredictable [49].

This study provides new information about patients’ experiences of periprosthetic knee infection. Saturation was achieved in the sample of 16 patients from six UK NHS orthopaedic centres, and we took care to ensure rigour in analysis through a double coding and team science approach. We acknowledge that it is possible that the inclusion of more patients from additional study centres may have elicited supplementary findings. We also acknowledge that a study limitation, common to all research which employs opt-in consent is that there is an inherent self-selection bias, however, achievement of saturation gives us confidence that the sampling was appropriate in quantity and breadth. The study was conducted in the UK context, but we suggest that the experience of infection after knee replacement resonates with other contexts, including social impact and need for care. Interviewing participants between two and 10 months after revision surgery may have introduced recall bias but this approach gave the study the opportunity to explore the longer-term impact of infection, treatment and recovery.

The impact of periprosthetic infection is wide-ranging, and research to date has not paid sufficient attention to the experiences of people who have this complication after knee replacement. Participants within this study described a disruption to their everyday behaviours, a loss of identity, growing dependency on families and social support networks, as well as uncertainty about their futures. As such we suggest that periprosthetic knee infection may be described as an assault on patients’ physical self, sense of self and life course. Further research into the impact of infection might take a longitudinal, prospective approach to explore recovery and change over time in more detail. Our findings lead us to suggest that clinicians in primary, secondary and community care should be supported to provide consistency in care, not only through conveying the importance and urgency of infection diagnosis, but also by being vigilant to the early warning signs and symptoms of periprosthetic infection. Indeed, early diagnosis of periprosthetic infection maximises the chance of prosthesis retention [50]. Future research could focus on patient preparedness for adverse outcomes after joint replacement, help-seeking in the event of periprosthetic infection, and clinician support in the early recognition of periprosthetic infection.