Archives of Orthopaedic and Trauma Surgery

, Volume 124, Issue 10, pp 699–701

The influence of preclinical anaemia on outcome following total hip replacement


    • Regional Orthopaedic UnitOur Lady’s Hospital
  • P. O. Grady
    • Regional Orthopaedic UnitOur Lady’s Hospital
  • A. M. Dolan
    • Regional Orthopaedic UnitOur Lady’s Hospital
Original Article

DOI: 10.1007/s00402-004-0754-6

Cite this article as:
Myers, E., Grady, P.O. & Dolan, A.M. Arch Orthop Trauma Surg (2004) 124: 699. doi:10.1007/s00402-004-0754-6



Preoperative anaemia confers a higher intraoperative and postoperative risk of complications on a patient. Preclinical anaemia is defined as a patient with laboratory indices of anaemia but without any of the symptoms. The patient population who undergo total hip replacement (THR) are elderly and are more likely to fall into this category. Our aim was to assess the effect of preclinical anaemia on a patient’s postoperative course.

Materials and methods

A prospective analysis of elective THRs performed over a 10-month period was carried out. Preoperative haematological indices were recorded and correlated with postoperative complication rates.


A total of 225 elective THRs were included in the study. Patients with preclinical anaemia on admission had a higher incidence of postoperative infection and transfusion (p<0.001) and a longer postoperative inpatient stay. Preoperative iron supplementation in patients with preclinical iron deficiency anaemia resulted in a reduction in transfusion requirements (p=0.00125).


Identification and treatment of patients with preclinical anaemia preoperatively may reduce postoperative infection and transfusion needs and result in a shorter inpatient stay.


AnaemiaTotal hip replacementTransfusionInfection


Patients who are anaemic prior to surgery have a higher incidence of intraoperative and postoperative morbidity and mortality than those who have a normal haemoglobin (Hb) measurement [4]. There is also a category of patients with preclinical anaemia, defined as Hb <12.5 g/dl in men and <11.5 g/dl in women. These patients have laboratory values indicating anaemia but no clinical symptoms. They can escape the clinician’s attention until they are admitted to hospital preoperatively. The patient population who undergo total hip replacement (THR) are elderly and are more likely to fall into this category. In 1995 approximately 250,000 elective THRs were performed in the USA, at an estimated cost of $5 billion [5].Department of Health statistics reveal that 2785 primary THRs were performed in the public health sector in the Republic of Ireland in 1999.

Our objectives were:
  1. 1.

    To identify those patients who had preclinical anaemia.

  2. 2.

    To identify the type of anaemia.

  3. 3.

    To determine the transfusion and infection rates associated with preclinical anaemia.

  4. 4.

    To calculate the length of hospital stay compared with those patients with a normal Hb measurement undergoing the same operation.


Patients and methods

We prospectively analysed 225 elective primary hip arthroplasties performed between September 1999 and July 2000. Preoperative full blood count and haematological indices were recorded and correlated with the transfusion rate, the postoperative complication rate and the length of hospital stay.

Transfusion in the postoperative period was guided by a transfusion protocol (Fig. 1). The decision to transfuse was accompanied by an assessment of the patient’s underlying health and the development of clinical symptoms.
Fig. 1

Summary of the postoperative transfusion protocol

For the purposes of this study, a patient was deemed to have an urinary tract infection (UTI) only if there was microbiological evidence of this and to have a respiratory tract infection (RTI) if there was clinical, radiological or microbiological evidence to support the diagnosis.

All patients had their operation performed by one of six consultant orthopaedic surgeons. A cemented Charnley prosthesis was implanted in 21% (47/225) of cases. The remaining 79% (178/225) had a cemented Exeter prosthesis implanted. Revision THRs were excluded from the study. All patients who had clinical symptoms of anaemia were also excluded from the study.

Chi-square analysis was performed for statistical comparisons. Significance was set at p<0.05.The statistical package employed was Stat view (Stat view for Windows version 5.0.1).


A total of 225 elective THRs was performed over a 10-month period. Thirty-five (15%) patients had preclinical anaemia. The mean Hb measurement in the anaemic group was 11.8 g/dl compared with a mean Hb measurement of 13.2 g/dl in the non-anaemic group. The mean age in the study group was 62 years and in the control group, 64 years. The male to female ratio in the anaemic group was 2:1 and in the non-anaemic group, 3:2. The mean ASA grade in both groups was 2.

The number of patients in each group, the incidence of postoperative blood transfusion, RTI, UTI and length of stay is shown in Table 1.
Table 1

Comparison of postoperative complications between study and control groups (UTI urinary tract infection, RTI respiratory tract infection)

Study group

Control group


Number of patients




Blood transfusion

25 (71%)

20 (10.5%)



10 (28%)

26 (14%)



5 (14%)

23 (12%)


Length of stay (days)




Patients in the anaemic group had a higher incidence of postoperative UTI (p=0.039) and RTI (14% vs 12%). The postoperative transfusion rate in the anaemic group was 71% compared with 10.5% in the non-anaemic group (p<0.001) (Fig. 2). The average length of stay in the anaemic group was 18 days compared with 11 days in the non-anaemic group.
Fig. 2

Transfusion requirements, postoperative complications (RTI respiratory tract infection, UTI urinary tract infection) and length of stay (LOS) in the study vs control group

The preclinical anaemia group was further subdivided based on the type of anaemia (Table 2). This revealed that 9% of patients admitted had preoperative iron deficiency anaemia. Only 19% of them had been previously investigated for this, and all of them were taking iron supplements on admission; in this subgroup of patients, the transfusion rate was 50% compared with 88% in those patients who had iron deficiency anaemia but were not taking iron supplements (p=0.00125) (Fig. 3).
Table 2

Subdivision of the types of preclinical anaemias

Normochromic normocytic

Hypochromic microcytic







7 (58%)

18 (86%)



4 (33%)

6 (29%)



2 (17%)

3 (14%)


Length of stay (days)




Fig. 3

Transfusion rate in the iron deficiency anaemia subgroup, comparing those patients who received iron supplements preoperatively with those who did not


In our series 15% of patients admitted for elective primary hip arthroplasty had preclinical anaemia. As expected, the transfusion rate in this cohort of patients was significantly higher than that found in those patients with a normal Hb level (p<0.001). This concurs with similar studies [3, 6, 7]. Interestingly, in the iron deficiency anaemia group, the incidence of an intraoperative or postoperative blood transfusion was significantly lower in those patients who were on iron supplements preoperatively (p=0.00125). This suggests that iron supplementation in patients with iron deficiency anaemia may protect against a fall in Hb in the immediate postoperative period. This supports the view of Andrews et al. [1] who conclude that anaemic patients scheduled for primary hip replacement benefit significantly from preoperative iron supplements over 4 weeks.

Study group patients had a higher incidence of postoperative UTI (p=0.039) and RTI. This supports the findings of the serious hazards of transfusion (SHOT) reporting system [9], which details serious morbidity due to the transfusion of allogenic blood. It has been suggested that the transfusion of allogenic blood is an independent predictor of an increased risk for the development of postoperative infection [8].Bierbaum et al. [2] reviewed 10,000 total joint arthroplasties and found an increased rate of chest infection and UTI in transfused patients compared with those who did not require a blood transfusion. It is possible that a number of the postoperative infections in the normochromic normocytic subgroup may represent reactivation of a previously undetected chronic infection.

The mean length of stay in the study group was 18 days compared with 11 days in the control group. Each day in hospital costs an additional €600 in the Republic of Ireland, so this has obvious cost implications. Other studies have also found that transfusion leads to an increase in the length of hospital stay [2, 4].

Of the 35 patients who had preoperative anaemia, 60% had iron deficiency anaemia, 34% had normochromic normocytic anaemia, and 6% had macrocytic anaemia. Iron deficiency anaemia represents the most treatable of these preoperative anaemias. Our data would support the view that its treatment preoperatively prevents postoperative morbidity.

It is clearly beneficial not to transfuse blood if possible; this is especially the case in an elective surgical setting where transfusion can lead to an increase in postoperative infection and length of inpatient stay.

Our results indicate that all patients with preclinical iron deficiency anaemia should receive iron supplements prior to their elective surgery as this results in a decrease in transfusion requirements with all its inherent risks.

The investigation and treatment of preoperative anaemia should be a priority in preparing the patient for surgery, and systems need to be put into place to identify these patients before admission to hospital. Ideally, a preoperative assessment clinic involving both surgical and anaesthetic teams would identify those patients at risk. A simple measure to identify patients at risk would involve performing a full blood count upon referral by their GP. This would result in appropriate investigation and treatment prior to elective surgery being performed.

Copyright information

© Springer-Verlag 2004