Home About Journal AHEAD OF PRINT Current Issue Back Issues Instructions Submission Search Subscribe Blog    

Users Online: 731 
Print this page  Email this page Small font sizeDefault font sizeIncrease font size 

ARTHROPLASTY Table of Contents   
Year : 2005  |  Volume : 39  |  Issue : 2  |  Page : 93-98
Postoperative pyrexia after arthroplasty - when to panic ?

Department Of Orthopaedics, PD Hinduja Hospital, Mumbai, India

Click here for correspondence address and email


Background: Success of arthroplasty is contingent on a clear understanding of the potential complications. Today with improved methods of sepsis control, incidence of joint sepsis has dropped to less than 2%. Despite this fever is still common in the early post-operative period.
Methods: We reviewed 184 consecutive hip and knee replacement surgeries for incidence and clinical significance of post-operative fever. The cases were followed up for a period of over 3 to 5 years. Temperature charts up to 6 th postoperative day and all investigations were reviewed to determine the cause of fever.
Results: Post operative fever was recorded after 82 procedures (44.6%). The average maximum temperature occurred on post­operative day 1 (98.9 o F). Only 2 TKR got infected. Incidence of fever was higher in TKR as compared to THR.
Conclusion: Post-operative fever is common and probably inflammatory. It is not an important predictive factor of joint infection. Work up for joint infection is not indicated unless other corroborative features are present. Aspiration of painful joint is a highly accurate for identifying an infection.

Keywords: Post-operative pyrexia; Arthroplasty.

How to cite this article:
Agarwala S, Jain D, Bhagwat A. Postoperative pyrexia after arthroplasty - when to panic ?. Indian J Orthop 2005;39:93-8

How to cite this URL:
Agarwala S, Jain D, Bhagwat A. Postoperative pyrexia after arthroplasty - when to panic ?. Indian J Orthop [serial online] 2005 [cited 2020 Feb 26];39:93-8. Available from:

   Introduction Top

Infection after total hip or knee arthroplasty is a major concern for the orthopedic surgeon. Metallic foreign bodies contribute to local sepsis by decreasing the quantity of bacteria necessary to establish infection and by permitting the pathogens to persist on the surface of the avascular material, sequestered from the circulating immune factors and systemic antibiotics [1] . 'Glycocalyx' coating on the implants protect bacteria from host defense and make more difficult the diagnosis of sepsis by traditional diagnostic techniques [2] . Staphylococcus aureus Scientific Name Search  and  Staphylococcus epidermidis Scientific Name Search unt for most of the infections [3],[4],[5],[6]

Prosthetic joint infection occurs in approximately 1.5%­2.5% of all primary hip or knee arthroplasties [7] . A lot of investigations have been advised and done from time to time to detect the infection early, which includes ESR, CRP, plasma viscosity (PV), aspiration of joint, X-ray, bone scan; and other clinical features include fever, joint pain, etc. Pyrexia is the least commonly talked about topic in the detection of post operative infection. Etiological background and the relation of postoperative fever to wound infection is still unclear. The number of publications related to this topic in general and in the specific field of orthopedics is quite limited [8] . Documentation of normal fever response after total knee and hip replacement is important to avoid unnecessary work up for sepsis and to provide justification for early discharge [9] . The objective of the present study was to find out the significance of postoperative fever and other laboratory parameters as an indicator of infection after total joint arthroplasty.

   Materials and methods Top

Patient selection : This is a retrospective analysis of consecutive primary joint replacement surgeries of hip or knee performed between July1997 to January 2000. Of these 16 patients positive for immunodeficiency virus, hepatitis B and C antigen and those who did not have a follow up record of 3 to 5 years were excluded from study. Final study group therefore comprises 184 joint replacement surgeries in 145 patients with a minimum follow up of three years. Of these 45 patients were operated for unilateral total hip replacement and 61 unilateral total knee replacements. Six patients had replacement of both the hips while 33 patients had bilateral knee replacement. In none of the patients the two joints were replaced simultaneously.

Preoperative assessment and surgical procedure: All patients were hospitalized one or two days prior to surgery. Basic investigations like complete blood counts, serum electrolytes, blood urea, serum creatinine, fasting blood sugar, ECG and X-ray chest, X-rays of the knee and hip were carried out. The first named author performed all the replacements in the same vertical laminar airflow theatre. To keep chances of wound sepsis to a minimum shaving of the operative site was done in the operation theatre just prior to surgery [10] or part prepared with chemical epilation. All but 5 surgeries were done under general anesthesia. Exsanguination tourniquet was used for all knee replacement surgeries. Water-resistant drapes and antibiotic (gentamycin) laden bone cement was used for all surgeries. The usual duration of surgery was one and half-hour. Perioperatively 2 gram of cefazolin was injected intravenously at the time of induction, 1 gram 8 hours postoperatively at night on the day of surgery and 1 gram on the coming morning at 8.00 A.M. All patients received 2500 IU low molecular weight heparin (dalteparin sodium-fragmin Pharmacia) preoperatively 1 hour before surgery, followed by the same 8 hours after surgery and 5000 IU every day for 6 post operative days as DVT prophylaxis. Postoperatively knees were immobilsed either in a plaster slab or Jones type bulky dressing for two days. Suction drains were left in place for forty eight hours or until drainage was less than fifty millimeters in eight hours [11] . Immobilization was routinely discontinued on the third postoperative day and physical therapy consisting of continuous passive motion (CPM) and active assisted range of motion exercises was begun at that time. Patients were discharged as soon as they were able to walk well with walker, usually by eighth to tenth day postoperatively.

The 8 hourly postoperative temperature chart of each patient for 7 days was reviewed. The maximum temperature recorded over previous 24 hours was noted for purpose of data analysis.The day of surgery was taken as day 0.Postoperative day 1 began at 8:00am the following day. Subsequent postoperative days began and ended at 8:00am. On postoperative 1, 4 and 6th day complete blood count, fasting blood sugar, blood urea nitrogen, serum creatinine estimation was carried out and recorded in all patients. The results of other investigations like urine culture, blood culture, venous Doppler if clinical suspicion of DVT and chest X-ray when done in patients who had fever were recorded. The joints which were painful, warm, stiff and tender had complete and differential cell counts, CRP and ESR done. If the above investigations were suggestive of infection and no improvement occurred in 2-3 days time then bone scan was done, the joint was aspirated and material sent for grams staining and culture and sensitivity. The appropriate antibiotics were administered according to culture and sensitivity.

Postoperative pyrexia was defined as an axillary temperature more than 98.6 degree F or 37 degree C on any or all of the 6 days after surgery [12] . For the purpose of analyzing the data all calculations were made with regards to the number of surgeries and not the number of patients. Any patient who underwent 2 joint replacements was counted as 2 surgical interventions.

Post operatively a deep infection was diagnosed and included in this report if the bacterial culture of a specimen of the aspirate from the joint was positive.

   Results Top

The demographic data of the patients divided into group A (with fever) and group B (without fever) is shown in [Table - 1]. The average age of the patients was 60.4 years (range from 21 to 90 years). There were 55 surgeries in males and 129 surgeries in females. Fifty seven underwent THR while 127 had TKR surgeries. Fifteen patients had diabetes mellitus; 30 had systemic hypertension and 4 were suffering from both. Postoperative fever was recorded after 82 surgeries (44.6%). Seventeen of 57(29.8%) THR surgeries had postoperative fever while 65 of 127 (51.2%) TKR surgeries recorded a postoperative fever.

The distribution of fever as per postoperative day is shown in [Figure - 1]. Usually the maximum temperature was recorded on postoperative day 1, but it was not unusual for the maximum recorded temperature to occur on day of surgery or day 2.The temperature showed gradual defervescence, recording normal by day 6.The highest recorded temperature was 102 degrees Fahrenheit in 5 interventions (2.7% of 184 cases). Only one of these had joint sepsis in the immediate postoperative period. Maximum number of surgeries recorded fever on day 1 with a continuous decline as day's progressed [Table - 2]. Fifteen surgeries (18.3% of 82) developed positive clinical or laboratory findings to explain the cause of fever [Table - 3]. Of these 15, two (1.09% of 184 cases) developed a joint infection during the early postoperative period. Both were TKR surgeries done in RA patients. One patient had fever >100 o Fahrenheit on all seven days (day 0 to 6th postoperative day). Both these patients had considerable pain and joint aspiration carried out in the operation theatre under all aseptic conditions showed growth of staphylococcus aureus in both.

When hip or knee replacements surgeries were combined for study purposes, no statistically significant differences were identified in the following variables studied in relation to pyrexia

  • Patients younger than 60 versus patients more than 60 years of age.
  • A diagnosis of osteoarthritis, avascular necrosis of femoral head, fracture neck of femur and ankylosing spondylitis. But a comparatively higher incidence was found in RA patients. Intra operative and postoperative blood loss of less than 500cc versus those with a loss of more than 500 cc.
  • Surgeries with septic complications and those without it.

Similar results were found for these subgroups when patients who had a total hip replacement and total knee replacement were analyzed independently. The only statistically significant factor was that the patients with TKR had a significantly higher incidence of fever on all days than those who underwent a THR (p value<. 001). However there was no difference of the temperature curve between the two [Figure - 2].

An increase in the white blood cell count is seen after surgery, both in patients who had postoperative fever and those who did not have fever. The patients with septic complications had virtually the same values and identical curve of rise and fall in white blood cell count as those with fever but without septic complications. None of the chest X­rays taken to determine the etiology of fever had changes to suggest pulmonary embolism, atelactasis, pleural effusion or congestive heart failure.

   Discussion Top

The present study brings out the fact that early postoperative fever after arthroplasty is common (44.6%) in the first 7 days of surgery and is of unknown etiology. Similar figures have been quoted and conclusions drawn by Than et al [8] , Shaw and Chug [9] and Kennedy et al [12] . From their study of 118 total knee arthroplasty patients Guinn et al came to the conclusion that postoperative fever is not necessarily a contraindication to discharge from inpatient care [10].

Concurrent infection as a cause of postoperative fever after joint replacement accounts for only a few cases. In our study, the probable cause could only be determined in 15 out of 82 patients [Table - 3]. Of these, joint infection was the cause of fever for 2 surgeries only (1.09% of 184 cases). Thus infection in general and in particular, joint infection is not the most common cause of postoperative fever. The only possible clue to joint infection was persistent and disabling joint pain. Grogan et al [11] and Russel [13] also found pain and not the fever as major presenting symptom in the joint sepsis.

The peak value of temperature does not correlate with the presence of infection as only 1 of the 5 patients with a fever of more than 102 0 F had joint sepsis. Marwin et al in a retrospective study of 295 patients of TKR had 80% of patients with a temperature of >101 0 F with only 6 % having postoperative infections[14] . Based on their data the positive predictive value (PPV) of temperature 101 0 F for any infection was 1% and 13% PPV of temp >102 0 .

Other clues to distinguish fever due to infection from that due to postoperative trauma show no statistical significant difference between the subgroups of age, sex or blood loss except for the fact that patients with TKR had an increased incidence of fever compared to THR(p value<. 05) and increased incidence in RA patients [12] . Both the knees which got infected were suffering from RA. Knees affected with RA are well documented of having higher incidence of infection [4],[6],[11] . Both were infected with staphylococcus aureus which is the most common infective organism [4],[7] . Patients with RA have increased predilection for infection due perhaps to a greater tendency towards delayed or failed wound healing and diminished host resistance, mainly because of decreased neutrophil function [4] .

The cause of fever after 67 surgeries (81.7%) of febrile cases could not be ascertained. This may be explained by the fact that typical postoperative fever is inflammatory in etiology and probably stems from leukocytic degradation of extravasated blood and other soft tissue and bone debris and pyrogenic materials released from injured host cells [15,[16],[17] . Macrophages, which are important in the development of the acute phase response, are common in bone and bone marrow, but occur less often in skeletal muscle. The degree of bone and marrow injury produced during THR and TKR may be important in determining the level of fever and increase in WBC count. TKR is reported to be a more traumatic procedure than THR and therefore induces a higher level of acute phase reactants. White et al claim this to be the reason for the greater increase in C-reactive proteins after TKR as compared to THR [18] . Husain and Kim [19] found that rise in CRP and ESR and Okafor and Stanmore [20] opined that ESR, CRP and plasma viscosity (PV) is an acute phase response. The term 'acute phase' refers to local and systemic events that accompany inflammation. Local response includes vasodilation, platelet aggregation, neutrophil chemotaxis and release of lysosomal enzymes. Systemic response includes fever, leukocytoses and change in a hepatic synthesis of acute phase proteins.

The ESR may be affected by size and shape of red blood cells, plasma composition and fluid status. CRP, in contrast, is independent of physical properties. If a rising trend, discordant from that of established pattern is observed- it is a biological warning and should raise an index of suspicion for infection [19],[20] . ESR reaches a peak on 5th day while CRP on 3 rd day [21] . There have been suggestions implicating hepatic cellular damage secondary to bone cement toxicity as a cause of rise in postoperative temperature [22] .

Fever due to wound infection generally becomes evident on the fourth to seventh postoperative day [23] . Acute leucocytosis occurs in stressful conditions with all types of surgery, general and epidural anesthesia due to rise in levels of endogenous epinephrine, nor-epinephrine and cortisol [24] . A higher leukocyte count (>35,000 mm cube) occurs as a result of suppuration anywhere in the body [25] . A rise in the WBC count in a patient with unresolving joint pain could be due to joint infection [Figure 4]. The endotoxin released by bacteria is thought to contribute to the production of high leukocyte count.

From this study we conclude that postoperative fever and leucocytosis after arthroplasty is a normal physiologic response attributable to degradation of tissue debris and extravasated blood [12] . They are not a reliable marker of joint infection [2],[13],[25] . In fact presence of fever is only a minor predictive factor of wound infection, while at the same time absence of fever does not mean absence of septic complications. As long as the maximum daily temperatures and leukocyte count shows a decreasing trend and there are no corroborating signs and symptoms of sepsis like increased joint pain, dysuria or breathlessness, postoperative fever can be considered to be an inflammatory reaction to surgery. Only if the temperature continues to be high after the fifth day or rises over 38 0 C (101 degree F) successively, it starts to be a source of concern. A workout for signs of postoperative infection, infection in the urinary tract, deep venous thrombosis, postoperative lung inflammation, haematoma (blood collecting in the operative wound), and incipient ossification in the muscles is then indicated[9],[10].

The patient's entire clinical picture should be considered when entertaining the diagnosis of infection. If excruciating joint pain is the overriding complaint [26] and swelling is present aspiration of painful joint is recommended, as it is a highly accurate method for identifying those joints that are infected [2],[25],[27] . We propose an algorithm to be followed in patients who have early postoperative fever after arthroplasty [Figure - 3].[28]

   References Top

1.Brause BD. Infected total knee replacement - diagnostic, therapeutic and prophylactic considerations. Orthop Clin North Am.1982;13: 245­49.  Back to cited text no. 1    
2. Cuckler JM , Star AM, Alavi A, Noto RB. Diagnosis and management of infected total joint arthroplasty. Orthop Clin North Am.1991; 22: 523-30.  Back to cited text no. 2    
3. Ayers DC, Dennis DA, Johanson NA, Pellegrini VD. Common com­plications of total knee arthroplasty. J Bone Joint Surg (Am). 1997; 79: 278-311.  Back to cited text no. 3    
4. Wilson MG, Kelley K, Thornhill TS. Infection as a complication of total knee replacement arthroplasty. J Bone Joint Surg (Am). 1990; 72: 878-83.  Back to cited text no. 4    
5. Masterson EL, Masri BA, Duncan CP. Treatment of infection at the site of total hip replacement. J Bone Joint Surg (Am). 1997; 79: 1740-49.  Back to cited text no. 5    
6. Petty W, Bryan RS, Coventry MB, et al. Infection after total knee arthroplasty. Orthop Clin North Am. 1975; 6:1005-14.  Back to cited text no. 6    
7. Lentino JR. Prosthetic joint infections: bane of orthopedists, challenge for infectious disease specialists. Clin Infect Dis. 2003 May 1; 36(9): 1157-61.  Back to cited text no. 7    
8. Than P, Malovics I. Significance of postoperative fever after hip prosthesis implantation. Zeitschrift fur orthopadie und thre grenzgebiete.2000,138(5): 430-5.  Back to cited text no. 8    
9. Shaw JA and Chun R: Febrile response after knee and hip arthro­plasty. Clin Orthop.1999, 367: 181-89.  Back to cited text no. 9    
10. Guinn S, Castro FP Jr., Garcia R, Barrack RL. Fever following total knee arthroplasty; Am J Knee Surg. 1999;12 (3):161-4.  Back to cited text no. 10    
11. Grogan TJ, Dorey F, Rollins J, Amstutz HC. Deep sepsis following total knee arthroplasty. J Bone Joint Surg (Am). 1986, 68, 226-234.  Back to cited text no. 11    
12. Kennedy G, Rogers WB, Rurakowiski D, et al: Pyrexia after total knee replacement: A cause for concern? Am J Orthop 1997, 26: 549- 554.  Back to cited text no. 12    
13. Windsor RE. Management of total knee arthroplasty infection. Orthop Clin North Am. 1991; 22: 531-537.  Back to cited text no. 13    
14. Marvin SE. The significance of fever in the immediate postoperative period after total knee replacement. Poster presentation at Am Acad­emy Orthop Surg: Feb 13-17, 2002. Dallas,TX.  Back to cited text no. 14    
15. Atkins E. Pathogenesis of fever. Physiology Rev. 1960, 40:580-646.  Back to cited text no. 15    
16. Guyton AC: Blood Groups; Transfusion; Tissue and Organ Transplan­tation. In: Guyton. Textbook of Medical Physiology. 4th ed. Philadelphia: WB Saunders. 1971, 126-135.  Back to cited text no. 16    
17. Guyton AC: Body Temperature, Temperature Regulation, and Fever. In: Guyton. Text book of medical physiology. 4th ed. Philadelphia: WB Saunders. 1971, 831-843.  Back to cited text no. 17    
18. White J, Kelley M, Dunsmuir R. C-reactive protein level after total hip and total knee replacement. J Bone Joint Surg (Br). 1998, 80: 909-11.  Back to cited text no. 18    
19. Husain TM, Kim DH. C-Reactive protein and erythrocyte sedimenta­tion rate in Orthopaedics. Univ Pennsylvania Orthop J. 2002;15: 13-16.  Back to cited text no. 19    
20. Okafor B, Stanmore MG. Postoperative changes of erythrocyte sedi­mentation rate, plasma viscosity and C-reactive protein levels after hip surgery. Acta Orthop Belg. 1998; 64(1):52-6.  Back to cited text no. 20    
21. Moreschini O, Greggi G, Giordano MC, Nocente M, Margheritini F. Postoperative physiopathological analysis of inflammatory param­eters in patients undergoing hip or knee arthroplasty. Int J Tissue React. 2001, 23(4):151-4.  Back to cited text no. 21    
22. Ritter MA, Gioe TJ, Sieber JM. Systemic effect of polymethacrylate. Increased serum levels of gammaglutamyltranspeptidase following ar­throplasty. Acta Orthop Scand. 1984, 55: 411-413.  Back to cited text no. 22    
23. Schawrtz SI. Complications. In: Principles of Surgery. 5th edition Vol. 1. Singapore: Mc Graw Hill , 1988; 469-497.  Back to cited text no. 23    
24. Dale DC. Neutrophilia- In: Willams Hematology -5 th edition. New York: Mc Graw Hill Inc, 1995; 825-828.  Back to cited text no. 24    
25. O'Neil DA, Harris WH. Failed total hip replacement: assessment by plain radiographs, arthrograms, and aspiration of the hip joint. J Bone Joint Surg (Am). 1984, 66: 540.  Back to cited text no. 25    
26. Merkel KD, Brown ML, Dewanjee MK, Fitzgerald RH. Comparison of Indium labeled leukocyte imaging with sequential Technetium-Gal­lium scanning in the diagnosis of low-grade musculoskeletal sepsis. J Bone Joint Surg (Am). 1985, 67.  Back to cited text no. 26    
27. Philips WC, Kattapurum SV. Efficacy of preoperative hip aspiration performed in the radiology department. Clin Orthop.1983;179: 141.  Back to cited text no. 27    
28. Heffernan J A, Puertas C PI, Aguirre R AC. 99m Tc labeled antigranulocyte antibody fragment Fab scintigraphy and three-phase bone scintigraphy in the study of painful hip and knee prosthesis. Rev Esp Med Nucl.2002; 21(4): 286-9.  Back to cited text no. 28    

Correspondence Address:
Sanjay Agarwala
PD Hinduja Hospital, Mahim, Mumbai -400 016
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/0019-5413.36781

Rights and Permissions


  [Figure - 1], [Figure - 2], [Figure - 3]

  [Table - 1], [Table - 2], [Table - 3]

This article has been cited by
1 Postoperative fever: A normal inflammatory response or cause for concern
Lindsay Burke
Journal of the American Academy of Nurse Practitioners. 2010; 22(4): 192
[VIEW] | [DOI]


    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Email Alert *
    Add to My List *
* Registration required (free)  

    Materials and me...
    Article Figures
    Article Tables

 Article Access Statistics
    PDF Downloaded333    
    Comments [Add]    
    Cited by others 1    

Recommend this journal