The Journal of International Medical Research
2001; 29: 7 – 12
C-reactive Protein Values and Erythrocyte
Sedimentation Rates after Total Hip and
Total Knee Arthroplasty
ÖF BILGEN1, T ATICI1, K DURAK1, O KARAEMINOĞULLARI2 AND MS BILGEN3
1Department of Orthopaedics and Traumatology, Uludağ University, Bursa, Turkey;
2Department of Orthopaedics and Traumatology, Abant Izzet Baysal University, Düzce,
Turkey; 3Department of Orthopaedics and Traumatology, Sarıyer Government Hospital,
Istanbul, Turkey
We studied the changes in serum C- on day 21 in THR patients and at the end
reactive protein levels (CRP) and of the second month in TKR patients. ESR
erythrocyte sedimentation rates (ESR) in peaked on day 5 after operation (THR
patients with primary osteoarthritis, who 100.5 mm/h, TKR 101.3 mm/h), dropping
underwent uncomplicated arthroplasty. Of close to pre-operative values at the end of
the 28 patients studied, 12 had cementless the third month in THR patients and at the
total hip replacement (THR), and 16 end of the ninth month in TKR patients,
underwent cemented total knee although, even after a year, ESRs were
replacement (TKR) with a tourniquet. In slightly above their pre-operative values.
both groups serum CRP levels increased Serum CRP levels changed more rapidly
rapidly after surgery, peaking on day 2 than ESRs and returned to normal more
(THR 23.17 mg/dl, TKR 26.02 mg/dl), and rapidly. CRP and ESR values tended to be
dropping gradually to pre-operative values higher in TKR than in THR patients.
KEY WORDS: C-REACTIVE PROTEIN; ERYTHROCYTE SEDIMENTATION RATE; ARTHROPLASTY
Introduction value between 89 mg/l and 135 mg/l 
(mean 115 mg/l).2 There were, for example,
Establishing a diagnosis of deep infection after statistically significant differences in plasma
arthroplasty is difficult, and early diagnosis CRP levels but not in ESRs before revision in
assists the choice of treatment and 
total hip prosthesis patients undergoing
the likelihood of a successful outcome.1 – 4
revisions because of mechanical loosening 
Investigations of plasma C-reactive protein
(CRP) levels and erythrocyte sedimentation (n = 23), compared with those whose
rates (ESR) in patients with infected revisions were needed because of septic
2
arthroplasty have produced varied results.2,5 – 9 loosening (n = 5). ESRs reached normal levels
In cases of infected arthroplasty, the within 3 – 6 months in 75 uncomplicated
sedimentation level may be between 16 mm/h total hip replacement (THR) patients but
and 62 mm/h (mean 35 mm/h) and the CRP always failed to reach the normal levels in 
7
ÖF Bilgen, T Atıcı, K Durak et al.
CRP values and ESRs after arthroplasty
18 patients with radiological and clinical To determine CRP and ESR levels, blood
findings of deep infection, and there was a samples were taken before breakfast pre-
statistically significant difference between the operatively, and post-operatively on days 1, 2,
ESRs of the two groups.5 In a study of 229 total 3, 5, 6, 7 and 21, and at the end of the first
knee replacement (TKR) patients and 301 THR second, third, sixth, ninth and twelfth months.
patients, there were no statistically significant Plasma CRP levels were measured quan-
differences in ESRs or CRP levels when 82 titatively by Behring® nephelometer (Behring
patients with wound infections (six of which Marburg GmbH, Marburg, Germany) and ESR
were deep infections) were compared with was determined by the Westergren method.
those without wound problems.4 In another We calculated mean percentage changes
study about a quarter of CRP levels and ESRs from the initial values for plasma CRP
were normal before revision in 23 THR concentrations and ESR. Statistical comparisons
patients in whom revisions were done because were done using the Wilcoxon test.
of deep infection.8
After uncomplicated arthroplasty, the Results
time taken for CRP levels and ESRs to reach The mean pre-operative CRP values in the
their peaks, and then to return to normal THR and TKR groups were equal at 0.54 mg/dl
levels seems to vary.1,2,6,10 – 13 In the present and the mean ESRs were 14.12 mm/h and
study, we investigated the timing of the 16.08 mm/h, respectively. The highest mean
changes in plasma CRP levels and ESRs in CRP values, 26.02 mg/dl in TKR patients and
patients with uncomplicated arthroplasty. 23.17 mg/dl in THR patients, occurred on day
Patients and methods 2 after operation (Fig. 1). Plasma CRP values
decreased thereafter to means of 1.0 mg/dl in
We studied 28 uncomplicated cases of THR patients (by day 21) and 0.88 mg/dl in
primary osteoarthritis, 12 of whom the TKR group (at the end of the second
underwent THR (43%) and 16 (57%) of month); these values did not differ
whom underwent TKR between January significantly from the pre-operative values 
1998 and May 1998. The THR group (P > 0.05). By the end of the third month the
comprised eight women and four men with a mean plasma CRP values in both groups were
mean age of 64 years and the TKR group, similar to the pre-operative values. There were
four men and 12 women with a mean age of statistically significant differences between
61 years. For infection prophylaxis a 1500-mg the mean plasma CRP values in THR and TKR
dose of cefuroxime was given intravenously patients on day 21 and at the end of the first
before and two 750-mg doses were given month (P < 0.05). There were no other
after the operation. For prophylaxis of deep statistically significant differences but mean
venous thrombosis low-molecular-weight CRP values in TKR patients were also higher
heparin (Dalteparine) 5000 IU was injected on days 1, 2, 3 and 7, and at the ends of the
subcutaneously pre-operatively and then second and third months. These differences
once daily for 2 weeks. Regional anaesthesia were slightly more marked when the changes
was used in all cases. In the THR group non- were calculated as percentages of the pre-
cemented prostheses were inserted using a operative value (data not shown).
posterolateral approach but in TKR patients Mean ESRs were at their highest, 100.5 mm/h
cemented prostheses were inserted using an and 101.3 mm/h in THR and TKR patients,
anterior approach and a tourniquet. respectively, on post-operative day 5 (Fig. 2).
8
ÖF Bilgen, T Atıcı, K Durak et al.
CRP values and ESRs after arthroplasty
30
25
20 Total knee replacement
Total hip replacement
15
10
5
0
Start 1 2 3 4 5 6 7 21 1 2 3 6 9 12
Day Month
Time after surgery
FIGURE 1: Mean plasma C-reactive protein concentrations in patients who underwent
total knee replacement () or total hip replacement (■)
The ESRs decreased thereafter to means of Diagnosis of infection after arthroplasty is
25 mm/h at the end of the third month in THR quite difficult. Besides clinical, radiological
patients and 25.6 mm/h in the ninth month in and scintillographic findings, laboratory
TKR patients, values that did not differ determinations of plasma CRP levels and
significantly from the pre-operative values (P > ESRs are valuable in the diagnosis of
0.05). The only significant difference between infection and in the selection of appropriate
the ESRs of the THR and TKR groups was on treatment.3,16 – 18 The CRP level reacts more
day 1 (P < 0.05). At the end of the twelfth rapidly and is more sensitive than is the
month the ESRs in THR and TKR patients were ESR.2,5,7,9,11,19,20 Further, the use of cement and
21.3 and 22.1 mm/h, respectively, higher than tourniquets in operations does not appear to
the pre-operative values. When the changes in affect CRP levels or ESRs.6,10,11,20,22
ESRs were calculated as percentages of the pre- Previous studies have reported that
operative value, higher percentage variations plasma CRP levels in THR and TKR patients
were found in the TKR group than in the THR increase rapidly after operation reaching
group (Fig. 3). maximum values within 2 – 3 days, after
which they decrease rapidly.1,6,10 – 12 Discussion In 186
patients undergoing uncomplicated
Plasma levels of acute-phase proteins, such cemented THR and TKR, plasma CRP
as CRP, synthesized by hepatocytes, and of reached a peak on day 2 after the operation
fibrinogen, which has an effect on the whereas, in 87 patients undergoing
acceleration of erythrocyte aggregation, uncemented THR and TKR, it reached its
increase in response to tissue trauma in peak on day 3.6 Plasma CRP levels peaked on
infection, inflammation and malignancy.14,15 day 3 in 109 THR patients, 97 of whom were
9
Plasma C-reactive protein concentration (mg/dl)
ÖF Bilgen, T Atıcı, K Durak et al.
CRP values and ESRs after arthroplasty
110
100
90
80
70 Total knee replacement
60 Total hip replacement
50
40
30
20
10
0
Start 1 2 3 4 5 6 7 21 1 2 3 6 9 12
Day Month
Time after surgery
FIGURE 2: Mean erythrocyte sedimentation rates in patients who underwent total
knee replacement () or total hip replacement (■)
800
700
600
500
Total knee replacement
400 Total hip replacement
300
200
100
0
1 2 3 4 5 6 7 21 1 2 3 6 9 12
Day Month
Time after surgery
FIGURE 3: Mean percentage variations in erythrocyte sedimentation rates in patients
who underwent total knee replacement () or total hip replacement (■)
10
% of pre-operative value Erythrocyte sedimentation rate (mm/h)
ÖF Bilgen, T Atıcı, K Durak et al.
CRP values and ESRs after arthroplasty
given cemented and 12 uncemented THR, plasma CRP levels in the TKR and THR
whereas in 39 patients undergoing groups occurred on day 21 and at the end of
unicondylar knee replacement (UKR), the first month, but plasma CRP values were
plasma CRP level peaked on day 2 after generally higher in TKR patients for the first
operation.10 In our group of patients plasma 3 months after surgery; CRP levels had
CRP levels reached their peak on day 2 after returned to pre-operative levels by day 21 in
operation. Cemented or uncemented THR patients and by the end of the second
application of prosthesis to a knee or hip month in TKR patients. These results suggest
joint does not seem to influence the changes that in TKR patients, compared with THR
in plasma CRP levels after operation (Fig. 1). patients, bone and bone marrow tissues were
In arthroplasty patients, the time taken for subjected to more trauma, and that in TKR
plasma CRP levels to return to normal after patients the degree of inflammation was
surgery has been reported as 2,13 3,2,10,11 or higher and the duration longer.
6 – 81 weeks, whereas White et al.12 reported In many studies, in THR and TKR patients,
that, at the first follow-up visit after discharge post-operative ESR has been reported to peak
from hospital, the CRP levels had returned to on days 5 – 7 and to decline after day 7.2,5,10,11
normal, although the exact time-period was The present findings that the ESR reached its
not reported. White et al.12 also reported that peak on day 5 after operation and began to
the post-operative CRP level in TKR patients decrease from day 6 are consistent with
was 50% more compared with that in THR earlier findings.5 The plasma CRP level
patients until day 7 after operation; reached its peak more rapidly than did the
thereafter the difference became statistically ESR, indicating that CRP levels are more
insignificant and plasma CRP returned to responsive, a finding consistent with the
normal levels at the same time in both literature.2,5,7,9,11,19,21 Statistically significant
groups. Niskanen et al.6 also reported that the differences between ESRs on day 1 after
timing of the return to normal plasma CRP surgery in our THR and TKR patients may 
levels was the same in TKR and THR patients, be attributable to haemodynamic changes
and found higher plasma CRP levels in TKR resulting from plasma and blood
patients than in THR patients. In our THR transfusions during THR operations, as
and TKR patients, the 3 months that the mentioned in earlier reports.2,11
plasma CRP level took to reach its normal In our study, ESRs at the end of the third
level was longer than the times previously month in THR patients and at the end of the
reported but, as previously reported, the ninth month in TKR patients did not differ
timing did not vary with technique. 6,12 In hip significantly from pre-operative values. This
and knee arthroplasty, macrophages, which shows that ESR tends to decrease slowly
are found in greater numbers in bone and compared with CRP, as reported in
bone marrow than in muscle tissue, provide arthroplasty10 and in THR patients.22 The later
an effective means of determining plasma decrease in ESR in TKR patients compared with
CRP levels. The severity of bone and bone THR patients is likely to be for the reasons
marrow trauma was also an influencing mentioned above in connection with the later
factor. Since TKR is more traumatic, the decrease of CRP levels in TKR patients. The post-
plasma CRP level is expected to be higher.10 operative normalization of ESR in THR patients
In our present study the only statistically may take 3 – 6 months,5 or 4 months in THR
significant differences between the mean and TKR patients.13 The slight elevation of ESR
11
ÖF Bilgen, T Atıcı, K Durak et al.
CRP values and ESRs after arthroplasty
in our patients at the end of a year is consistent a year after surgery in some THR and 
with previous findings that ESR was high even UKR patients.2,10,11
• Received for publication 15 August 2000 • Accepted 21 August 2000
©2001 Cambridge Medical Publications
References level after total hip and total knee replacement. 
1 Choudhry RR, Rice RP, Triffitt PD, Harper WM, J Bone Joint Surg Br 1998; 80: 909 – 911.
Greg PJ: Plasma viscosity and C-reactive protein 13 Kolstad K, Levander H: Inflammatory
after total hip and knee arthroplasty. J Bone laboratory tests after joint replacement surgery.
Joint Surg Br 1992; 74: 523 – 524. Ups J Med Sci 1995; 100: 243 – 248.
2 Shih LY, Wu JJ, Yang DY: Erythrocyte 14 Covey C, Albright AJ: Clinical significance 
sedimentation rate and C-reactive protein of the erythrocyte sedimentation rate in
values in patients with total hip arthroplasty. orthopaedic surgery. J Bone Joint Surg Am 1987;
Clin Orthop 1987; 225: 238 – 246. 69: 148 – 151.
3 Peltola H, Vahvanen V, Aalto K: Fever, C- 15 Fischer CL, Gill C, Forrester MG, Nakamura R:
reactive protein and erythrocyte sedimentation Quantitation of ‘acute-phase proteins’
rate in monitoring recovery from septic postoperatively. Value in detection and
arthritis. J Pediatr Orthop 1984; 4: 170 – 174. monitoring of complications. Am J Clin Pathol
4 Gaine WJ, Ramamohan NA, Hullin MG, 1976; 66: 840 – 843.
McCreath SW: Wound infection in hip and knee 16 Gelman MI, Coleman RE, Stevens PM, Davey
arthroplasty. J Bone Joint Surg Br 2000; 82: BW: Radiography, radionuclide imaging and
561 – 565. arthrography in the evaluation of total hip 
5 Carlsson AS: Erythrocyte sedimentation rate and knee replacement. Radiology 1978; 128:
in infected and non-infected total hip 677 – 682.
arthroplasties. Acta Orthop Scand 1978; 49: 17 Tehranzadeh T, Schneider R, Freiberg RH:
287 – 290. Radiological evaluation of painful total hip
6 Niskanen RO, Korkala O, Pammo H: Serum replacement. Radiology 1981; 141: 355 – 362.
C-reactive protein levels after total hip and 18 Lyons CW, Berquist TH, Lyons JC, Rand JA,
knee arthroplasty. J Bone Joint Surg Br 1996; 78: Brown ML: Evaluation of radiographic findings
431 – 433. in painful hip arthroplasties. Clin Orthop 1985;
7 Foster IW, Crawford R: Sedimentation rate in 195: 239 – 251.
infected and uninfected total hip arthroplasty. 19 Kaperonis AA, Michelsen CB, Askanazi J,
Clin Orthop 1982; 168: 48 – 52. Kinney JM, Chien S: Effects of total hip
8 Sanzen L, Sundberg M: Periprosthetic low-grade replacement and bed rest on blood rheology
hip infections erythrocyte sedimentation rate and red cell metabolism. J Trauma 1998; 28:
and C-reactive protein in 23 cases. Acta Orthop 453 – 457.
Scand 1997; 68: 461 – 465. 20 Maury CPJ, Teppo AM, Raunio P: Control of the
9 Okafor B, MacLellan G: Postoperative changes acute phase serum amyloid A and c-reactive
of erythrocyte sedimentation rate, plasma protein response: comparision of total
viscosity and C-reactive protein levels after hip replacement of the hip and knee. Eur J Clin
surgery. Acta Orthop Belg 1998; 64: 52 – 56. Invest 1984; 14: 323 – 328.
10 Larsson S, Thelander U, Friberg S: C-reactive 21 Rosas MH, Leclercq S, Pegoix M, Darlas Y,
protein (CRP) levels after elective orthopedic Aubriot JH, Rousselot P, et al: Contribution of
surgery. Clin Orthop 1992; 275: 237 – 242. laboratory test, scintigraphy, and histology to
11 Aalto K, Österman K, Peltola H, Rasanen J: the diagnosis of lower limb joint replacement
Changes in erythrocyte sedimentation rate and infection. Rev Rhum Engl Ed 1998; 65: 477 – 482.
C-reactive protein after total hip arthroplasty. 22 Wroblewski BM: ESR and polymethyl-
Clin Orthop 1984; 184: 118 – 120. methacrylate. J R Coll Surg Edinb 1974; 19:
12 White J, Kelly M, Dunsmuir R: C-reactive protein 182 – 185.
Address for correspondence
Dr ÖF Bilgen
Department of Orthopaedics and Traumatology, Uludağ Universitiy, 
16000 Bursa, Turkey.
12