Skip to main content
NCBI home page
The BMJ logoLink to The BMJ
. 1998 Apr 4;316(7137):1089.

Intravascular volume optimisation during repair of proximal femoral fracture

Intravascular volume was depleted perioperatively in control group

KW Toh 1, W J Fawcett 1
PMCID: PMC1112910  PMID: 9558995

Editor—In their study investigating intravascular volume optimisation and proximal femoral fracture, Sinclair et al have a different approach to fluid replacement between the two groups studied.1 The control group received only a median volume of 1000 ml of crystalloid (presumably Hartmann’s or 0.9% saline solution) peroperatively. As only about a quarter of this solution is retained intravascularly, these patients effectively had a depleted intravascular volume perioperatively. In contrast, the protocol group received a median of 750 ml of colloid as well as 725 ml of crystalloid, giving intravascular fluid replacement of nearly 1 litre. The preoperative dextrose-saline maintenance fluids would have had minimal effect on the intravascular volume: only 15% of such a solution is retained within the intravascular compartment.

This study merely shows that patients who have not received fluid or who have fluid depletion have a higher rate of complications, which is not a surprising or new finding. A control group should be recognisable as such; if the control group has been deprived of fluid then this will magnify any benefits in the protocol group. We suggest that if adequate preoperative intravascular fluids, and not just dextrose-saline maintenance fluids, had been given to all patients, replacing the blood loss associated with femoral fractures (often in excess of 1 litre2), then the differences between the groups would have been much less. A study comparing a group in whom optimisation is used with a group given sufficient fluid would be of much greater importance and would show the benefits of optimisation more clearly.

References

  • 1.Sinclair S, James S, Singer M. Intraoperative intravascular volume optimisation and length of hospital stay after repair of proximal femoral fracture: randomised controlled trial. BMJ. 1997;315:909–912. doi: 10.1136/bmj.315.7113.909. . (11 October.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Willet KM, Dorrell H, Kelly P. ABC of major trauma: Management of limb injuries. BMJ. 1990;301:229–233. doi: 10.1136/bmj.301.6745.229. [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ. 1998 Apr 4;316(7137):1089.

Differences in outcome were probably due to chance

Martyn J Parker 1

Editor—Sinclair et al’s randomised controlled trial of intraoperative intravascular volume optimisation in patients with proximal femoral fracture raises important issues regarding the preoperative administration of fluid to such patients and their operative care,1-1 particularly if it is taken in conjunction with a study by Schultz et al.1-2 Sinclair et al’s study, in only 40 patients, suggested that optimisation of the intraoperative stroke volume resulted in a large reduction in hospital stay but not mortality. Schultz et al’s study, in 70 patients, showed a reduction in mortality, from 10 of 35 patients in the control group to 1 of 35 patients in the group who had intensive physiological monitoring. Hospital stay was not reported on.

Results of both trials must be interpreted with more caution than that shown in the editorial by Gan and Arrowsmith.1-3 It is difficult to explain why each study produced profound changes in dissimilar outcome measures. Both studies were of small numbers of patients. Patients with hip fracture are a heterogeneous group, and multiple factors affect outcome. It is therefore unlikely that a change in one aspect of patient care would have such a large influence on the eventual outcome and much more plausible that the differences in outcome are due to chance.

A large number of observational studies have generally failed to suggest that isolated differences in anaesthetic methods, operative procedures, or preoperative care have significant effects on length of hospital stay or mortality. Gan and Arrowsmith’s comment—based on only 40 cases—that routine use of oesophageal monitoring during surgery for hip fracture will save 450 000 hospital bed days a year in Britain alone is misleading and unhelpful in the critical appraisal of an innovation.

Undoubtedly we need further research, on a large number of patients in different clinical settings, to see if these potential benefits of new technology can be translated into real benefits. Until such studies are conducted, the role of intensive monitoring must be regarded as promising but unproved.

References

  • 1-1.Sinclair S, James S, Singer M. Intraoperative intravascular volume optimisation and length of hospital stay after repair of proximal femoral fracture: randomised controlled trial. BMJ. 1997;315:909–912. doi: 10.1136/bmj.315.7113.909. . (11 October.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 1-2.Schultz RJ, Whitfield GF, LaMura JJ, Raciti A, Krishnamurthy S. The role of physiologic monitoring in patients with fractures of the hip. J Trauma. 1985;25:309–317. doi: 10.1097/00005373-198504000-00005. [DOI] [PubMed] [Google Scholar]
  • 1-3.Gan TJ, Arrowsmith JE. The oesophageal Doppler monitor: a safe means of monitoring the circulation. BMJ. 1997;315:893–894. doi: 10.1136/bmj.315.7113.893. [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ. 1998 Apr 4;316(7137):1089.

Regional anaesthesia is usually technique of choice

Donal Buggy 1, Gerard Fitzpatrick 1

Editor—We wish to make several comments about Sinclair et al’s randomised controlled trial on intraoperative intravascular volume optimisation and length of hospital stay after repair of proximal femoral fracture.2-1

Firstly, regional anaesthesia is generally accepted to be the technique of choice in most patients with traumatic fractures of the neck of femur, because of improved outcome in general and a reduced incidence of postoperative thrombotic complications in particular.2-2 Sinclair et al, however, specifically excluded patients for whom regional anaesthesia was planned because use of this technique would preclude the placement of an oesophageal probe. Surely this fact minimises the impact of their findings, as patients having general anaesthesia for repair of proximal femoral fracture are a relatively small subgroup of this population.

Secondly, the only haemodynamic data provided by the authors are stroke volume, cardiac output, and aortic flow time. While cardiac output is one of the main factors influencing mean arterial blood pressure and hence organ perfusion, the other main factor contributing to mean arterial pressure—systemic vascular resistance—is not reported. The potential beneficial effect of increased cardiac output on organ perfusion may be negated by large reductions in systemic vascular resistance.

Sinclair et al’s findings of improved cardiac output when colloid was given to patients under general anaesthesia conflict with studies in patients undergoing spinal anaesthesia, in which cardiac output was measured non-invasively by transthoracic bioimpedance. In one such study, prehydration with gelatin colloid did not prevent the reduction in blood pressure induced by spinal anaesthesia in normovolaemic elderly patients, whereas an infusion of the α-adrenergic agonist metaraminol did.2-3

Our study comparing prehydration with gelatin colloid, prehydration with crystalloid, and no prehydration found no benefit, in terms of haemodynamic stability, with either fluid regimen over none.2-4 In a further investigation, hetastarch 6% given in combination with crystalloid was also found to be ineffective compared with intramuscular methoxamine (another α-adrenergic agonist) in an elderly population with fractures of the femoral neck.2-5 In neither of these studies, however, did we undertake the postoperative follow up of outcomes conducted by Sinclair et al.

Further studies, in which similar postoperative outcome measures are used as end points, are needed in patients undergoing spinal anaesthesia, who constitute the overwhelming majority having this surgery. Such studies must determine whether similar benefits may be obtained by close control of cardiac output and other haemodynamic variables, especially systemic vascular resistance, by using supplementary colloids or vasopressors.

References

  • 2-1.Sinclair S, James S, Singer M. Intraoperative intravascular volume optimisation and length of hospital stay after repair of proximal femoral fracture: randomised controlled trial. BMJ. 1997;315:909–912. doi: 10.1136/bmj.315.7113.909. . (11 October.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2-2.Modig J, Borg I, Bagge L, Saldeen T. Role of epidural and general anaesthesia in fibrinolysis and coagulation after total hip replacement. Br J Anaesth. 1983;55:625–629. doi: 10.1093/bja/55.7.625. [DOI] [PubMed] [Google Scholar]
  • 2-3.Critchley L, Conway F. Hypotension during subarachnoid anaesthesia: haemodynamic effects of colloid and metaraminol. Br J Anaesth. 1996;76:734–736. doi: 10.1093/bja/76.5.734. [DOI] [PubMed] [Google Scholar]
  • 2-4.Buggy D, Higgins P, Moran C, O’Brien D, O’Donovan F, McCarroll M. Prevention of spinal anesthesia-induced hypotension in the elderly: comparison between preanesthetic administration of crystalloid, colloid and no prehydration. Anesth Analg. 1997;84:106–110. doi: 10.1097/00000539-199701000-00020. [DOI] [PubMed] [Google Scholar]
  • 2-5.Buggy DJ, Power CK, Meeke R, O’Callaghan S, Moran C, O’Brien R. Prevention of spinal anesthesia-induced hypotension in the elderly: intramuscular methoxamine or prehydration with 6% hetastarch and crystalloid combined. Br J Anaesth. 1998;80:199–202. doi: 10.1093/bja/80.2.199. [DOI] [PubMed] [Google Scholar]
BMJ. 1998 Apr 4;316(7137):1089.

Authors’ reply

Mervyn Singer 1,2, Sue Sinclair 1,2

Editor—We agree with Toh and Fawcett that our control patients were under-resuscitated and fluid deplete. In our patients, initial resuscitation, which included blood, was conducted as clinically appropriate. All patients then received preoperative crystalloid during the period of obligatory starvation, a practice that is not common in many hospitals. Furthermore, by a Hawthorn effect, more attention would probably have been paid to peroperative circulatory status in our control patients.

We would argue that optimisation is impossible without flow monitoring. To answer Parker, is it not ironic that the two interventional studies that monitored and acted on flow data (ours and that by Schultz et al3-1) have been the only studies that have shown significant outcome benefit in surgery for hip fracture? Of course, this could be due to chance. The large number of observational studies that Parker mentioned have failed to consider the impact of adequate fluid resuscitation. We hope that our study has highlighted this neglected area.

Parker is unfair to criticise Gan and Arrowsmith’s editorial, which stated the potential gain through flow monitoring but argued for large scale confirmatory studies before its routine introduction.

We and others3-2 would dispute Buggy and Fitzpatrick’s claim that regional anaesthesia is the technique of choice in surgery for hip fracture. A meta-analysis has not shown improved outcome.3-2 Their quoted reference from a 1983 paper concerns total hip replacement, not traumatic fracture. Surgery for hip fracture under general anaesthesia may be relatively uncommon in their hospital but is unsubstantiated by any general data. In the 1993-4 national confidential enquiry into perioperative deaths,3-3 23% of the deaths followed surgery for hip fracture and only a minority of the patients received regional anaesthesia alone. We are confused by their interpretation of physiology. A large reduction in systemic vascular resistance is relevant only if blood pressure and organ perfusion pressure drop below a critical level. At the end of a sprint a runner’s systemic vascular resistance will be 40% of normal but he or she does not generally require vasoconstrictors. Heart rate and blood pressure did not change significantly in either group in our study.

Finally, the use of vasopressors in elderly patients is not without danger. Flow may be severely compromised, especially with coexisting hypovolaemia. Flow monitoring is necessary both to confirm benefit and to exclude harm in individual patients. We would encourage the authors to perform outcome studies to determine whether the subset of patients with compromised flow after administration of an α agonist develop complications in the postoperative period.

References

  • 3-1.Schultz RJ, Whitfield GF, LaMura JJ, Raciti A, Krishnamuthy S. The role of physiologic monitoring in patients with fractures of the hip. J Trauma. 1985;25:309–316. doi: 10.1097/00005373-198504000-00005. [DOI] [PubMed] [Google Scholar]
  • 3-2.Sorenson RM, Pace NL. Anesthetic techniques during surgical repair of femoral neck fractures. A meta-analysis. Anesthesiology. 1992;77:1095–1104. doi: 10.1097/00000542-199212000-00009. [DOI] [PubMed] [Google Scholar]
  • 3-3.Report of the national confidential enquiry into perioperative deaths (NCEPOD) 1993/1994. London: NCEPOD; 1996. [Google Scholar]

Articles from BMJ : British Medical Journal are provided here courtesy of BMJ Publishing Group

RESOURCES