Joint contracture following prolonged stay in the
intensive care unit
Abstract
Background: Prolonged immobility during a critical illness may predispose patients to the development of joint contracture. We sought to document the incidence of, the risk factors for and the reversibility of joint contractures among patients who stayed in a tertiary intensive care unit (ICU) for 2 weeks or longer.
Methods: We conducted a chart review to collect data on the presence of and risk factors for joint contractures in the shoulders, elbows, hips, knees and ankles among patients admitted to the ICU between January 2003 and March 2005.
Results: At the time of transfer out of the ICU, at least 1 joint contracture was recorded in 61 (39%) of 155 patients; 52 (34%) of the patients had joint contractures of an extent documented to impair function. Time spent in the ICU was a significant risk factor for contracture: a stay of 8 weeks or longer was associated with a significantly greater risk of any joint contracture than a stay of 2 to 3 weeks (adjusted odds ratio [OR] 7.09, 95% confidence interval (CI) 1.29–38.9; p = 0.02). Among the variables tested, only the use of steroids conferred a protective effect against joint contractures (adjusted OR 0.35, 95% CI 0.14–0.83; p = 0.02). At the time of discharge to home, which occurred a median of 6.6 weeks after transfer out of intensive care, 50 (34%) of the 147 patients not lost to follow-up still had 1 or more joint contractures, and 34 (23%) of the patients had at least 1 functionally significant joint contracture.
Interpretation: Following a prolonged stay in the ICU, a functionally significant contracture of a major joint occurred in more than one-third of patients, and most of these contractures persisted until the time of discharge to home.
Joint contracture is a limitation in the passive range of motion of a joint secondary to shortening of the periarticular connective tissues and muscles.1–7Immobility plays a major role in the development of joint contractures.1,2,6,8–13Indeed, patients with conditions limiting mobility are at high risk for joint contracture.1,9,14,15 Prolonged immobility from critical illness can also be expected to predispose patients to experience joint contractures. Herridge and colleagues13 evaluated the long-term consequences of acute lung injury and noticed the disabling effects of joint contractures in a small number of patients in the first year after discharge from an acute care hospital. In a systematic review, we were unable to identify any other studies characterizing joint mobility after critical illness.16
Given the potentially devastating consequences of joint contracture after a prolonged stay in the intensive care unit (ICU), we sought to document the incidence of this problem among patients who remained in an ICU for 2 weeks or more. We were especially interested in mild, as well as functionally significant, limitation in the range of motion of major joints and in the number of joints affected. We also examined risk factors and the persistence or reversibility of joint contractures until discharge from hospital.
Methods
Study design and participants
We conducted a retrospective evaluation of hospital records for consecutive patients admitted to the ICU of a tertiary care hospital between January 2003 and March 2005. We included patients who stayed in the ICU for 14 days or more and who survived until transfer out of the unit. We retrieved records from the hospital's ICU database and extracted all charts that met our inclusion criteria. No patients were eliminated during the course of the study. The protocol for this project was approved by the Ottawa Hospital Research Ethics Board, The Rehabilitation Centre Ethics Board and the SCO Health Services Research Ethics Board.
Data collection and outcomes
We examined all chart entries for medical care, physiotherapy and occupational therapy. We extracted bilateral numeric range-of-motion data for 5 large joints: the shoulder, elbow, hip, knee and ankle. We tabulated the range of motion for the left and right sides, as well as 2 directions of movement for each joint, except the ankle, for which we extracted only lack of dorsiflexion. Diagnosis of a joint contracture required documentation of a measurement of range of motion; a progress note stating “patient had a contracture” or “bilateral ankle contracture” was not counted.
On the basis of the existing literature, we defined contracture in 2 ways. First, we defined “any contracture” as a recorded range of motion that was short of the full range.17–21 This definition reflects the fact that the activities of people with physical occupations and recreational or performance athletes will be limited by even small limitations in the range of motion of key joints.22–25 It also allowed us to quantify a potential dose–effect relation between exposure to the ICU and severity of limitation in the range of motion. In addition, we defined “functionally significant contracture” as more severe limitation in the range of motion, to an extent that has been established in the literature as causing functional limitation8,17,19,21,26 (Table 1). We recorded any contractures, functionally significant contractures, patients with at least 1 contracture, number of joint contractures per patient and the total number of joints with contractures.
Table 1.
For each patient, we sought range-of-motion data at 2 time points: at transfer out of the ICU and before discharge to home (patient's home, a nursing home or a regional hospital). Range of motion before discharge was recorded from the last medical unit before discharge to home.
In addition, we collected data on patient demographic characteristics, comorbidities, admission diagnosis, Acute Physiology and Chronic Health Evaluation (APACHE II) severity score27 on admission, length of stay in the ICU, duration of invasive mechanical ventilation, use of co-interventions (specifically, neuromuscular blockade and steroids) and total length of the hospital stay.
Statistical analysis
Patient demographic characteristics were summarized as frequencies and percentages for categorical variables. Continuous data are displayed as mean (standard deviation [SD]).
Two separate multiple logistic regression analyses were undertaken to evaluate the odds of experiencing any contracture and a functionally significant contracture, both determined at the time of transfer out of the ICU and before discharge to home. All independent variables that were considered clinically relevant by the study investigators were included in both analyses, specifically age (< 45, 45–65, > 65 years), sex, presence of diabetes mellitus, admission diagnoses (acute or chronic respiratory disease, cancer, neurologic or vascular disease, or sepsis), APACHE II severity score (< 15, 15–25, > 25), length of stay in the ICU (2–2.99, 3–4.99, 5–7.99, ≥ 8 weeks), duration of invasive mechanical ventilation (< 10, 10–19, ≥ 20 days), use of neuromuscular blockade or steroids as co-interventions (given v. not given) and length of hospital stay (< 4, 4–7.99, ≥ 8 weeks).
Results
In total, 155 patients met the inclusion criteria (Figure 1). Their mean age was 59.6 [SD 15.5] years (Table 2). The mean length of stay in the ICU was 3.1 (95% confidence interval [CI] 2.4–5.0) weeks. While in the ICU, at least 1 contracture was reported for 61 (39%) of the patients, and 52 (34%) of the patients had a contracture that met the criteria for functional significance.
Figure 1: Flow diagram for patient recruitment. Discharge home means discharge to the patient's home, a nursing home or a regional hospital.
Table 2.
At the time of transfer out of the ICU, a total of 212 major joints representing 61 patients were affected by any joint contracture, of which 144 joints (68%) were affected by a functionally significant contracture (Table 3). The elbow was the joint most frequently affected by any contracture (76 [35.8% of the total number of joints affected]), followed by the ankle (51 [24.1%]), the knee (31 [14.6%]), the hip (30 [14.2%]) and the shoulder (24 [11.3%]) (Table 3).
Table 3.
Each affected patient had an average of 3.5 joint contractures at the time of transfer out of the ICU, of which 2.4 contractures were functionally significant.
Time in the ICU was a significant risk factor for the development of any joint contracture. Among patients who remained in the ICU for 8 weeks or more, the adjusted odds ratio of experiencing any joint contracture was 7.09 (95% CI 1.29–38.9; p = 0.02) relative to those who stayed for 2 to 3 weeks (Figure 2). Similarly, the adjusted OR of experiencing a functionally significant contracture after 8 weeks or more in the ICU was 5.79 (95% CI 1.08–31.0; p = 0.04) relative to a stay of 2 to 3 weeks (Figure 3). Receiving steroids while in the ICU was associated with lower odds of developing any joint contracture (OR 0.35, 95% CI 0.14–0.83; p = 0.02) (Figure 2). The other variables tested (age, sex, admission diagnosis, diabetes mellitus as a comorbidity, duration of invasive mechanical ventilation and use of neuromuscular blockers) did not affect the risk of experiencing any joint contracture or a functionally significant joint contracture (Figure 2 and Figure 3).
Figure 2: Adjusted odds of experiencing any joint contracture. The results are based on a single multivariable model simultaneously including the demographic variables age and sex, presence of diabetes mellitus as a comorbidity, admission diagnosis, Acute Physiology and Chronic Health Evaluation (APACHE II) severity score, length of stay in the intensive care unit, duration of invasive mechanical ventilation, co-interventions (specifically neuromuscular blockade and steroids) and length of stay in the hospital. For definitions of variables see Table 2. Note: ref = reference group.
Figure 3: Adjusted odds of experiencing a functionally significant joint contracture. The results are based on a single multivariable model simultaneously including the demographic variables age and sex, presence of diabetes mellitus as a comorbidity, admission diagnosis, Acute Physiology and Chronic Health Evaluation (APACHE II) severity score, length of stay in the intensive care unit, duration of invasive mechanical ventilation, co-interventions (specifically neuromuscular blockade and steroids) and length of stay in hospital. For definitions of variables see Table 2. Note: ref = reference group.
After transfer out of the ICU, 8 patients were sent to a regional hospital and were lost to follow-up (Figure 1). The other patients were discharged to home directly from the ICU (n = 6), from the hospital ward (n = 96) or from a rehabilitation unit (n = 45); the median time until discharge to home after transfer out of the ICU was 6.6 weeks.
Upon discharge home, 50 (34%) of the 147 patients still had at least 1 joint contracture, and 34 (23%) of the 147 patients had at least 1 functionally significant contracture (Table 3). Of the 182 joint contractures that were still present at the time of discharge to home, 90 were functionally significant (Table 3). Among the 50 affected patients not lost to follow-up who had at least 1 contracture at the time of discharge to home, the mean number of contractures at that time was 3.6 per patient, and the mean number of functionally significant contractures was 1.8.
Interpretation
More than one-third of patients who stayed for 2 weeks or longer in the ICU had a functionally significant joint contracture. On average, these patients each had more than 2 functionally significant joint contractures. The duration of the ICU stay was associated with the presence of any joint contractures at the time of transfer out of intensive care. At 2 weeks, these patients were already experiencing functionally significant joint contractures. Thereafter, relative to patients who stayed in the ICU for just 2 to 3 weeks, the adjusted OR was greater among those who stayed for 5 to 8 weeks and was even higher among those who stayed for 8 weeks or more. Neither sex nor age played a role in the occurrence of any joint contractures in this population. The rate of joint contracture among patients who were admitted to intensive care for a neurologic or vascular disease and among those who received neuromuscular blockade was similar to that of other patients. Conversely, receiving steroids protected against any joint contractures.
One of the most important findings was the potential persistence of joint contractures until the time of discharge home. Our data indicate that joint contractures did not resolve spontaneously before discharge. About one-quarter of the patients were discharged to home with, on average, 2 joint contractures severe enough to interfere with daily activities.8,17,19,21–26
Several previous reports have mentioned persistent functional deficits after immobility.2,5,8,15,16,28,29 Joint contractures have been identified as a potentially important cause of such deficits, but their prevalence and risk factors have not been quantified.13,16 The severity in the restriction of range of motion and the location of a joint contracture determine the patient's functional limitations. Upper limb contractures impair tasks such as showering, dressing, feeding, hand to occiput and hand to perineum;15,26 lower limb contractures affect ambulation5,8,10,28,30 and increase the risk of falls.30 Of the 39% of patients in our study who had any contractures, most (85%) had 1 or more contractures with severity that was considered functionally important. For patients with physical occupations and for athletes in virtually all sports, a loss of even a few degrees of the full range of motion may be detrimental to performance.22–25
Contracture at a single joint may induce compensatory strategies. Contractures of multiple joints compound the patient's difficulties in performing activities of daily living and leisure pastimes and impose a burden on health care providers and on family members.5,8 Contractures of normal joints subjected to immobility are theoretically preventable. In this tertiary care ICU, 1.25 full-time physiotherapists and a 0.75 full-time occupational therapist (with assistance from nurses for positioning) were responsible for 24 intensive care beds. However, our data suggest that these usual monitoring and preventive activities were insufficient to prevent the joint contractures.
These data concur with findings of alterations of fibrosis,7 synovial shortening31 and decreased synoviocyte proliferation32 in the joint capsule of rats, measured as early as 2 weeks after immobilization of a normal joint and continuing for the next 30 weeks. The dose–response relation between exposure to intensive care and development of joint contracture supports immobility as a key pathophysiologic risk factor for joint contractures, without a supplemental contribution from affected neurologic systems. Given that multiple contractures were present, immobility related to multiple local causes (e.g. arterial or venous catheters, restraints, pressure sores) or generalized immobility can be incriminated. Our results are consistent with experimental literature showing that exogenous or endogenous steroid hormones improve the elasticity of periarticular soft tissues.33–35
Our study had limitations. Our range-of-motion data relied on the chart entries of health care professionals with different measuring and charting patterns. We analyzed only quantitative entries; cases with non-numeric, qualitative reporting of contractures and cases with no report of contracture (where contracture might have been present but went unrecorded) were defined as no contracture. This limitation would lead to underestimation of the true incidence of joint contractures. Baseline range of motion was unavailable, so some patients might have had one or more joint contractures before admission to the ICU, which would lead to overestimation of the incidence of contractures. Furthermore, not all directions or all joints were assessed for all patients. Data were not extracted for extension, external rotation, internal rotation or adduction of the shoulder; pronation–supination of the elbow; abduction, adduction, internal rotation or external rotation of the hip; or flexion or inversion–eversion of the ankle. Small joints such as the wrist, temporomandibular joint, and the joints of the hands, fingers, feet, toes and neck were not included in our study. The limited number of range-of-motion directions and the limited number of joints studied would lead to underestimation of the incidence of joint contractures. Finally, the retrospective design did not allow us to question patients to confirm the limitations associated with joint contracture we defined as functionally significant. Rather, we based our definition on authoritative literature involving measurement of functional limitations based on the range of motion of individual joints.
Our findings imply that patients requiring a lengthy stay in the ICU should be monitored and treated to prevent the appearance of joint contractures that could persist long after the patient is sent home. The lack of complete reversibility at discharge to home that we observed suggests that the natural evolution of joint contractures is not benign and that an expectation of spontaneous recovery could lead to increases in disability in this patient population. These data also underscore the difficulty of treating established joint contractures. Usual hospital care and rehabilitation in a large Canadian academic urban centre were insufficient to reverse the contracture process.
Our study suggests that prevention of joint contractures should be considered a central issue for critical care patients, akin to prevention of thromboembolic events and stress ulcers. As a rule, patients are admitted to the ICU because of major organ failure, not as a result of joint problems. However, prolonged immobility of normal joints predisposes critically ill patients to the development of contractures. In our study, many patients who were saved from life-threatening illnesses left the hospital with contractures severe enough to cause functional impairment. Joint contractures acquired in the ICU may lead to substantial costs associated with increased length of stay in hospital; increased need for rehabilitation treatments, outpatient treatments and use of devices for activities of daily living or gait; increased need for personal care at home; and loss of productivity due to inability to return to work.
Conclusions
Thirty-nine percent of patients staying for 2 weeks or more in an academic hospital ICU experienced joint contractures, often in more than 1 joint. Most of these contractures were of an extent documented to impair function and persisted until the time of discharge to home. This study indicates a need for prospective multicentre observational studies to further study joint contracture in the intensive care setting. The magnitude of the clinical problem also invites concerted actions to better understand the cause of contractures and develop effective interventions.
@ See related article page 725
Footnotes
- Une version française de ce résumé est disponible à l'adressewww.cmaj.ca/cgi/content/full/178/6/691/DC1Disclaimer: Paul Hébert is a coauthor of this article. As Editor-in-Chief ofCMAJ, he was not involved in the selection of peer reviewers or the vetting of the manuscript before its acceptance.This article has been peer reviewed.Contributors: Heidi Clavet, Paul Hébert, Dean Fergusson and Guy Trudel contributed to the conception and design of the project. Heidi Clavet drafted the article and was responsible for acquisition of the data. Steve Doucette contributed to the analysis and interpretation of the data and drafted the text describing the data analysis. Guy Trudel contributed substantively to the analysis and interpretation of the data. Paul Hébert, Dean Fergusson and Guy Trudel revised the article for important intellectual content. All of the authors gave final approval of the version to be published.Acknowledgement: The preliminary results of this study were presented at the annual congress of the Canadian Association of Physical Medicine and Rehabilitation, June 13–16, 2007, London, Ontario.Competing interests: None declared.
Nenhum comentário:
Postar um comentário