Systemic medical complications following joint replacement: a review of the evidence

Abstract

Introduction

Arthroplasty procedures are commonly performed in the UK. Informed consent is required for each procedure. To obtain informed consent the patient and their surgeon should discuss the risks and benefits of the proposed operation. This discussion should include both regional and systemic complication rates. Regional complications of arthroplasty are generally well documented in the literature. Systemic medical complications are less well described. This lack of accurate data could make it difficult for the treating surgeon to obtain valid consent. The aim of this paper was to review and compare the literature regarding the rate of systemic medical complications after common arthroplasty procedures.

Methods

A literature search was conducted using the PubMed, Cochrane Library and MEDLINE databases. Studies regarding the systemic medical complications and mortality rate of joint replacement were included.

Findings

We found that systemic complications were more frequent than regional complications following arthroplasty. The systemic complication rates were: hip, 5.1%; knee, 6.9%; ankle, 3.0%; shoulder, 11.2%; elbow, 8.5%; and wrist, 0%. Mortality rates for arthroplasty procedures were: hip, 0.3%; knee, 0.2%; ankle, 0.3%; shoulder, 0.3%; elbow, 0.2%; and wrist, 0%.

Conclusions

The most common systemic medical complication following arthroplasty was venous thromboembolism. Preoperative comorbidity was the most important risk factor for both postoperative mortality and systemic medical complications following arthroplasty procedures. We recommend that to obtain informed consent the given rates of systemic medical complications of joint replacement should be discussed and documented.

Introduction

The United Kingdom National Joint Registry showed increasing numbers of arthroplasty procedures performed between 2003 and 2019.¹ The most common procedures were total hip and knee replacement of which a total of 160,000 were performed each year.¹

For each arthroplasty procedure informed consent was required. The consent process should include a discussion involving the patient and the surgeon of the risks and benefits of the procedure. The likely complications, both regional and systemic, should be discussed before a shared decision is reached.² Operative complications may be defined as ‘any undesirable or unintended event that occurs to the patient as a direct result of the operation which would not have occurred had the operation gone as well as could reasonably be hoped’.³

To obtain consent, the surgeon will rely on the current literature. Most of the literature focuses upon regional complication rates such as wound infection and wound dehiscence.^4,5 Systemic complications are less well described. This lack of data regarding systemic complications may result in inadequate consent.

The aim of this review was to summarise the available literature regarding the systemic complications and mortality rate of common arthroplasty procedures.

Methods

Search strategy

A literature search was conducted using the PubMed, Cochrane Library and MEDLINE databases. Appendix 1 describes the search strategy performed. Backward chaining of reference lists from retrieved papers was also used to expand the search.

Inclusion criteria were studies regarding the systemic medical complications and mortality rate of joint replacement. Systemic medical complications were defined as all acute medical pathologies in the peri- and postoperative periods. Any studies focusing purely on regional complications, not in the English language or with no full text available were excluded. A time limit of 10 years (2011–2021) was used when performing the search strategy to ensure inclusion of relevant contemporaneous data that encompassed the development of current enhanced recovery protocols and multidisciplinary preoperative anaesthetic assessments routinely undertaken for most major joint arthroplasties.^6,7 These were developed with the aims of improving patient outcomes and reducing complications through rigorous preoperative optimisation.

Findings

Overall rate of systemic medical complications

Table 1 demonstrates the overall rate of systemic medical complications for common arthroplasty procedures. This was between 3% and 11.2% for most joints. We found that the systemic medical complications of hip and knee replacement were well documented in the literature. In comparison, the reported rates of other common arthroplasty procedures were underreported; most notably for wrist arthroplasty for which no systemic complications were reported.⁵

Table 1 .

Overall incidence of systemic medical complications after arthroplasty procedures

Joint	Systemic medical complication rate (%)	Reference	Sample size
Hip	5.1 (4.3–5.8)	Belmont et al4	17,640
		Pulido et al8	7,699
Knee	6.9 (5.0–8.8)	Belmont et al9	15,321
		Pulido et al8	5,818
Ankle	3.0	Curlewis et al10	40,353
Shoulder	11.2 (8.9–13.4)	Cancienne et al11	22,968
		Craig et al12	58,054
Elbow	8.5	Toor et al13	3,184
Wrist	0	Eckhoff et al5	57

Mortality rate

Table 2 demonstrates the mortality rate after arthroplasty procedures. For most procedures was between 0.2% and 0.3%. The literature was difficult to interpret due to the varying length of follow-up. Common periods of follow-up reported were 30 days, 90 days, the duration of inpatient stay or follow-up not stated.

Table 2 .

Postoperative mortality rate following arthroplasty procedures

Joint	Mortality rate (%)	Reference
Hip	0.3 (0.2–0.4)	Belmont et al4
		Pulido et al8
Knee	0.2	Belmont et al9
		Pulido et al8
Ankle	0.3	Curlewis et al10
Shoulder	0.3 (0.1–0.5)	Anakwenze et al14
		Boddapati et al15
		Craig et al12
Elbow	0.2	Toor et al13
Wrist	0	Eckhoff et al5

We identified age, preoperative comorbidities and revision arthroplasty as risk factors for mortality in arthroplasty procedures.

Age

Patient age >80  years was associated with significantly increased risk of mortality after hip and knee arthroplasty (p<0.001).^4,9 Following shoulder arthroplasty, age >84 years was associated with a higher mortality rate of 0.28%.¹⁶ Age had no significant effect on mortality rate after ankle, elbow or wrist arthroplasty.^5,10,13

Preoperative comorbidities

Patients with preoperative comorbidities (such as renal insufficiency or diabetes) had increased mortality rates following hip, knee and shoulder arthroplasty.^4,9,14 Preoperative comorbidities did not influence the mortality rate for ankle, elbow or wrist arthroplasty.^5,10,13

Revision arthroplasty

The incidence of in-hospital mortality was significantly greater in patients undergoing a revision procedure for hip and knee arthroplasty (p<0.05).⁸ This association was not found for other joint arthroplasty procedures.

Systemic medical complications of joint replacement by body system

Data concerning systemic medical complications by body system were limited (particularly for the central nervous and gastrointestinal systems) and significantly heterogenous, which made extrapolating this information and reporting exact percentages difficult.

The most common body systems affected across all joint arthroplasty procedures in rank order were as follows.

• •Cardiovascular system (CVS), including venous thromboembolism (VTE). Systemic medical complications involving the CVS were the most common.^{4,5,8–10,12,13} VTE was the most common systemic medical complication overall (1.12%–3.0%).^8,11 Other, less-common CVS complications reported included myocardial infarction and arrythmia (0.5%–1.53%).^8,12
• •Respiratory system. Hospital-acquired pneumonia (HAP) was the most common respiratory system complication (1%–2%).^4,5,8–13 Other respiratory complications, such as atelectasis, respiratory failure and pneumothorax were uncommon (<1%).^4,8,9,12
• •Genitourinary system (GUS). Acute kidney injury and postoperative urinary tract infection (UTI) were the most reported GUS complications (0%–2%).^4,8,9,12 Other GUS complications, such as urinary retention, were rare (0.07%).⁸
• •Central nervous system. Stroke was the sole identified central nervous system complication.^4,8–10,12
• •Gastrointestinal system. Postoperative ileus was the only reported gastrointestinal system complication.^4,8–10,12

Comparison of the regional and systemic complication rates

The regional complication rates for arthroplasty procedures are well documented in the literature (Table 3) and are typically classified as either major or minor. The most common major regional complications across all arthroplasty procedures were deep wound infection and peripheral nerve injury.^{4,5,8–10,12,13} The most common minor regional complications across all arthroplasty procedures were superficial wound infection and wound dehiscence.^{4,5,8–10,12,13}

Table 3 .

Regional complication rates after arthroplasty procedures

Joint	Common regional complications	Rate (%)	Reference
Hip	Deep wound infection	0.51	Belmont et al4
	Superficial wound infection	0.83
	Wound dehiscence	0.14
	Peripheral nerve injury	0.11
Knee	Deep wound infection	0.30	Belmont et al9
	Superficial wound infection	0.79
	Wound dehiscence	0.27
	Peripheral nerve injury	0.10
Ankle	Deep wound infection	0–5.7	Clough et al17
	Superficial wound infection	1.3–3.8
Shoulder	Wound infection	0.6	Floyd et al18
Elbow	Periprosthetic joint infection	1–12.5	Kwak et al19
Wrist	None stated	N/A	Eckhoff et al5

Risk factors for systemic medical complications

Several variables were assessed regarding their relationship to systemic medical complications after joint replacement.

Age

Age >80 years was found to be a strong predictor of developing a systemic medical complication in hip, knee and shoulder replacement surgery.^4,8,9,11,12 There was conflicting evidence in the literature regarding age and systemic medical complications following total ankle arthroplasty (TAA), with no strong relationship demonstrated in the literature.¹⁰ There was no data on the influence of age systemic medical complications on elbow¹³ or wrist⁵ arthroplasty.

Sex

Females were at higher risk of systemic medical complications after shoulder arthroplasty.^12,20 There was inconsistent evidence on the impact of sex on systemic medical complications after all other arthroplasty procedures. For example, Belmont et al⁴ found that males were at greater risk of major systemic complications after hip arthroplasty (p<0.01) and females were at greater risk of minor systemic complications after knee arthroplasty (odds ratio [OR] 1.22, 95% confidence interval [CI] 1.01–1.48).⁹ Pulido et al⁸ found no relationship between sex and systemic complication in hip and knee arthroplasty in a large single-centre prospective cohort study.

Obesity

Higher body mass index (BMI) was found to be an independent risk factor for systemic medical complications after TAA.¹⁰ BMI was not found to influence systemic medical complications after other arthroplasty procedures.^{4,5,8,9,11–13}

Systemic comorbidities

Systemic comorbidities, including an American Society of Anesthesiologists (ASA) classification of 3 or higher was associated with an increased risk of systemic medical complications in hip (p<0.0001),⁴ knee (OR 1.81, 95% CI 1.41–2.31)⁹ and shoulder arthroplasty.¹⁵ In patients undergoing TAA, any type of medical comorbidity was associated with an increased risk of deep venous thrombosis and infection (such as UTI and aspiration pneumonia).²¹ Increasing numbers of preoperative comorbidities were associated with a significant risk of developing at least one complication within 30 days of surgery.²¹

Diabetes mellitus

Patients with diabetes mellitus had an increased rate of pneumonia (relative risk [RR] 2.6), UTI (RR 1.9) and stroke (RR 9.1) after total elbow arthroplasty.¹⁹ Patients with diabetes mellitus who underwent TAA were more likely to experience pulmonary embolism (p<0.01), stroke (p<0.003), UTI (p=0.05) and overall complications (p<0.02) compared with patients without diabetes mellitus.²² Major systemic complications were more common in diabetic patients who underwent hip arthroplasty (p<0.002)⁴ and knee arthroplasty (OR 1.36, 95% CI 1.03–1.81).⁹ No association was stated in the literature for wrist arthroplasty.⁵

Perioperative blood transfusion

From the identified studies, a relationship between perioperative blood transfusion and increased systemic medical complications was only identified in shoulder and ankle arthroplasty.

Patients receiving a blood transfusion following shoulder arthroplasty had significantly higher rates of myocardial infarction, HAP and sepsis in the first 7 days postoperatively.²³ They also suffered significantly higher rates of acute renal failure, arrhythmia, bleeding complications, deep venous thrombosis, pulmonary emboli, heart failure, myocardial infarction, HAP, respiratory failure, sepsis/systemic inflammatory response syndrome (SIRS), stroke and UTI within 30 days and 90 days.²³ After ankle replacement, patients who received blood transfusion were significantly more likely to suffer from congestive heart failure, peripheral vascular disease, hypothyroidism, coagulation disorder, anaemia and acute renal failure.²⁴

Operative time and duration of anaesthesia

There was a positive correlation between length of operation and systemic medical complications for hip, knee, shoulder and ankle replacement.^4,8–12 This relationship was not investigated in the identified literature for elbow¹³ or wrist replacement.⁵

Operations lasting 141 min or more for hip arthroplasty (p<0.0001)⁴ and 135 min for knee arthroplasty (OR 1.52, 95% CI 1.19–1.93)⁹ meant that patients were at greater risk of major systemic medical complications. Anaesthesia time >200 min was associated with increased perioperative adverse events in patients undergoing a TAA (OR 2.83, 95% CI 1.10–7.28).²⁵

Discussion

We found the rate of systemic medical complications following arthroplasty to be between 3.0% and 11.2% for many of the joints studied. The highest rate was following shoulder replacement (11.2%). The lowest rate was for wrist replacement (0%), but few data were available for wrist arthroplasty, with only one study identified with a small cohort of 57 patients.⁵ As such, results for this relatively low volume arthroplasty (compared with hip, knee or shoulder arthroplasty) may be skewed owing to the small cohorts reported.

We were unable to be certain why shoulder arthroplasty had a higher systemic complication rate compared with other joints. The following factors may all have contributed: the process of data collection; the combination of a general anaesthetic, regional block and operative site all in close proximity; and the fact that many shoulder surgeons and anaesthetists do not undertake high numbers of shoulder arthroplasty procedures (relative to hip and knee surgeons volumes) leading to a longer learning curve and a higher complication rate.

The most affected body systems were the cardiovascular, respiratory and genitourinary systems. VTE was the single most reported systemic medical complication for all joint replacement procedures. Overall, systemic medical complications were found to be more common than regional complications.

We found the mortality rate of arthroplasty procedures to be between 0.2% and 0.3% for most of the joints assessed. Few data were available for wrist replacement and a rate of 0% was recorded in the literature.

To reduce the complication and mortality rates we suggest that patients with risk factors for postoperative complications (the elderly, those with diabetes and those with a high ASA) are identified and optimised prior to surgery. Their consent process ought to reflect their increased risk of complications. The operative time and length of anaesthesia for arthroplasty procedures should be reduced as much as safely possible.

Limitations

This review had several limitations. First, the available literature was significantly heterogenous, which made comparison between studies difficult. Second, many of the studies identified were retrospective cohort studies. This study design may fail to capture up to 50% of available data resulting in under-reporting of the real complication rate of the respective arthroplasty procedures.²⁶ Third, this review may have failed to capture all relevant papers. Although arthroplasty procedures have increased since 2003,¹ a time limit of 10 years was used for the search strategy to ensure that only relevant data with accurate complication rates according to modern surgical practises were included. To account for developments in surgical practice, we plan to repeat our review in the future so contemporaneous data is available to arthroplasty surgeons.

Conclusions

In conclusion we recommend that during the consenting process for joint arthroplasty of the hip, knee, shoulder, elbow and ankle the following risks are made clear to the patient and documented: a systemic complication rate of up to 11.2%; that the cardiovascular, respiratory and genitourinary systems are the most affected organ systems; and a mortality rate of 0.2%–0.3%.