SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements and conflicts of interest
  7. References

It remains uncertain whether spinal anaesthesia is preferable to general anaesthesia for surgical repair of hip fracture, but one determining factor is the comparative cost. A detailed cost analysis relating to 20 consultants’ intended anaesthetic practice (which provided information of consumables used) and data from the Brighton Hip Fracture Database was performed to quantify any difference in the costs of administering spinal versus general anaesthesia for patients with hip fracture. Although spinal anaesthesia took significantly longer to administer (mean (SD) time 31 (15) min vs 27 (16) min; p < 0.0001), the mean (SD) cost of spinal anaesthesia (£193.81 (37.49)) was significantly less than the cost of general anaesthesia (£270.58 (44.68); p < 0.0001). The mean percentage cost of anaesthesia was 3.8% of hospital income per hip fracture, and personnel contributed approximately 46% of this cost. While such considerations indicate that spinal anaesthesia is financially preferable, it is unknown whether differential clinical outcomes between regional and general anaesthesia may offset this apparent monetary advantage.

Approximately 77 000 patients require surgery for proximal femoral fracture annually in the UK [1–4], a figure projected to rise to approximately 95 000 by 2030 as the population ages [5]. For similar demographic reasons, it has been estimated that the incidence of proximal femoral fracture worldwide will rise from 1.7 million (1990) to 6.3 million by 2050 [6].

In 2000, the total UK hospital inpatient expenditure per hip fracture was estimated to be £4760 [7], equivalent to £225 million nationally per annum. In 2005, this was estimated to have risen to £12 163 (equivalent to £864 million nationally per annum), the rise related to increased length of hospital stay [8]. In 2008, a study from the Republic of Ireland calculated the mean total hospital expenditure per patient as approximately £8000 (with ward costs contributing 55%, operative costs 40% and investigations 5% of this figure), suggesting that the costs of this operation (although different in different countries) are generally increasing over time [9]. Inpatient costs account for 50% of the total cost of a fall related injury, with additional costs accruing through general practice, outpatient and rehabilitation attendance, and loss of earnings amongst carers [10].

Recently published research by the National Health Service Hip Fracture Anaesthesia Network has confirmed earlier observations that the anaesthetic technique varies widely for hip fracture repair. The lack of consensus on best practice results from a lack of research evidence [11–14].

The cost of spinal anaesthesia compared with general anaesthesia is relevant to the choice of anaesthetic technique. If both methods are equally effective and yield similar outcomes, then the least expensive is rationally preferable. For knee and hip arthroplasty, one previous study reported that the total cost per case (without personnel costs) was twice as much for general compared with spinal anaesthesia [15]. The main aim of this study was to assess if this was also the case when personnel costs were included for hip fracture repair.

Methods

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements and conflicts of interest
  7. References

In January 2009, we compiled a list of all anaesthetic equipment and drugs available to anaesthetists at the Royal Sussex County Hospital, Brighton, UK, calculating the cost per item using our central theatre stores catalogue and verifying these against company price lists and the British National Formulary[16].

With unit prices deleted, a complete equipment and drug list was distributed to 20 anaesthetic consultants who were asked to indicate the number of units of each item (for example, three electrocardiogram sensors, etc.) that they would routinely use when delivering either spinal anaesthesia (one list) or general anaesthesia with or without supplemental nerve blockade (a separate list) during a hypothetical, uncomplicated anaesthetic for hip fracture repair. As this was a service evaluation, specific consent was not sought, although it was explained that completion of the price list indicated consent to participation, and that final publication of results would omit any participant-identifiable information.

Consultants were asked to indicate the flow rate at which they administered oxygen, air and nitrous oxide both in the anaesthetic room and in theatre, and the percentage of inhalational agent that they would generally use during these phases of anaesthesia.

Once all 20 consultants had responded, data were transcribed to a Microsoft Excel 2003 spreadsheet (Microsoft Corporation, Redmond, WA, USA). The number of items used was multiplied by the unit cost per item, summation of which provided the cost of each mode of anaesthetic.

Medical gas costs were calculated per litre. The volume of gas derived from the liquid gas at standard temperature (20 °C) and pressure (101 kPa) was calculated (for example, 860 l of gaseous oxygen from 1 l of liquid oxygen). This figure was divided by the cost per litre (personal communication, British Oxygen Company (BOC), Guildford, UK: ∼ £0.001 for oxygen and air, ∼ £0.003 for nitrous oxide).

Inhalational anaesthetic usage (ml) was calculated using the Narkup computer model [17], assuming a patient’s weight of 55 kg, cardiac output of 4.5 l.min−1, minute ventilation of 4.5 l.min−1 and functional residual capacity of 1500 ml. This value was divided by the volume of a bottle refill for that agent and multiplied by the cost per bottle.

Both total gas and inhalational agent usage, and therefore cost, required estimation of the time spent in the anaesthetic room and in surgery. This information was derived from the Brighton Hip Fracture Database, which records, amongst other data, the time the patient entered the anaesthetic room to the time the patient entered theatre (‘anaesthetic time’), and the subsequent time until skin closure (‘surgical time’). Mean values for anaesthetic and surgical time were calculated from all operations performed exclusively under either spinal or general anaesthesia (with or without nerve block). We estimated that general anaesthesia would be administered during only 60% of the anaesthetic room time (after cannula insertion, and preoperative checks etc.), and factored this in to the calculations.

Anaesthesia personnel costs assumed one consultant in anaesthesia and one Operating Department Practitioner (ODP) per operation. Hourly wages were calculated from annual gross salaries – approximately £100 000 for consultants (48 h.week−1, £40.06 h−1), £25 000 for a mid-grade (Band 6) ODP (40 h.week−1, £12.02 h−1), and multiplied by the summated mean length of theatre episode (anaesthetic time plus surgical time) plus an estimated 15 min for pre-operative assessment (for consultant anaesthetists) and theatre set-up (for ODPs), and transfer to the postoperative care unit.

The mean Payment by Results hospital income per hip fracture repair was calculated by summing all historical 2006 Health Resource Group codes assigned by the hospital to OPCS 4.3 coded hip fracture procedures W461 (Primary prosthetic replacement of head of femur using cement), W471 (Primary prosthetic replacement of head of femur uncemented) and W191 (Primary open reduction of fracture of head of neck of femur & open reduction using pin & plate), and dividing the total by the number of procedures.

Statistical comparison of the length of anaesthetic and surgical times and cost of regional anaesthesia versus general anaesthesia was performed using Student’s t-test, with significance denoted by a p value < 0.05.

Results

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements and conflicts of interest
  7. References

All consultants provided spinal anaesthesia for hip fracture surgery (two of whom stated that they never administered general anaesthesia for hip fracture surgery such as hemi-arthroplasty or dynamic hip screw insertion).

From the Brighton Hip Fracture Database, the mean (SD) ‘anaesthetic’ time for patients undergoing surgery was 31 (15) min for 531 patients receiving only spinal anaesthesia compared with 27 (16) min (p < 0.0001) for 397 patients receiving general anaesthesia (with or without nerve block). The mean (SD) ‘surgical’ time was 76 (24) min compared with 81 (35) min (p = 0.01), respectively. Personnel costs were therefore calculated for either 122 min of spinal anaesthesia (i.e. 31 + 76 + 15 min) or 123 min (i.e. 27 + 81 + 15 min) of general anaesthesia, and gas/inhalational agent costs were calculated for 95 (19 + 76) min of spinal anaesthesia and 97 (16 + 81) min of general anaesthesia.

Equipment, drug, gas and personnel costs by anaesthetic type are shown in Table 1. Anaesthetic personnel costs were £0.87 min−1. The mean (SD) cost of inhalational general anaesthesia was £0.08 (£0.05) min−1.

Table 1.   Costs of hip fracture surgery by type of anaesthesia. Values are mean (SD).
 Anaesthesia equipment; £Airway equipment; £Personnel; £Drugs; £Gases/inhalational agents; £Total; £
  1. *p < 0.0001

Spinal anaesthesia66.73 (30.05)1.81 (0)105.90 (0)19.03 (11.00)0.43 (0.13)193.81 (37.49)*
General anaesthesia108.15 (38.53)25.68 (2.28)106.76 (0)25.17 (11.04)6.26 (3.94)270.58 (44.68)*

Table 2 shows the Payment by Results tariff received for 483 operations performed at the Royal Sussex County Hospital during 2005–2006, and the calculated cost of general and spinal anaesthesia as a percentage of revenue.

Table 2.   The calculated costs of general and spinal anaesthesia (absolute and as a percentage of revenue) for hip fracture surgery. Values are number (Propotion) or number.
 Cemented hemi-arthroplastyUncemented hemi-arthroplastyDynamic hip screwMean/Total
Mean PbR tariff/case, 2005–2006; £6713663051055910
GeneralSpinalGeneralSpinalGeneralSpinalGeneralSpinal
  1. PbR, Payment by Results

Number (N) of cases during study period43 (46%)51 (54%)55 (35%)101 (65%)109 (47%)124 (53%)207 (43%)276 (57%)
Revenue accrued during study period (× PbR tariff); £288 659342 363364 650669 630556 445633 0201 209 7541 645 013
Anaesthetic cost of procedure (× cost of anaesthesia from Table 1); £11 6359 88414 88219 57529 49324 03256 01053 492
Anaesthetic cost of procedure as a proportion of total PbR revenue4.0%2.9%4.1%2.9%5.3%3.8%4.6%3.3%

Overall, the reduction in cost associated with spinal anaesthesia can be expressed in several ways. In absolute terms there is saving of ∼ £80 per case (Table 1). Based on a mean operating room expenditure of £12–20 min−1 [18–20], we estimate that administering spinal compared with general anaesthesia would save 3.2–5.3% of the operating room cost. Or, expressed as a proportion of income per case, the estimated saving using spinal anaesthesia is ∼ 1.3% (Table 2).

Discussion

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements and conflicts of interest
  7. References

This study shows that spinal anaesthesia for proximal femoral fracture fixation is somewhat cheaper than general anaesthesia. Personnel costs were the most expensive item of expenditure, contributing approximately 46% of the cost of anaesthesia. Spinal anaesthesia took just ∼ 4 min longer to administer than general anaesthesia, but total theatre times were similar at ∼ 107 min, possibly due to emergence time from general anaesthesia.

We acknowledge that this study is limited by the sources of error inherent in financial, questionnaire, and retrospective database analyses. We took care to record ‘catalogue’ prices excluding local financial arrangements and bulk discounts, which may have led to a small overestimation of drug and equipment costs. We accept that our cost analysis of a ‘hypothetical’ anaesthetic is less accurate than a direct audit of equipment usage, although database analysis of 798 anaesthetics given by the 20 consultants over a 4-year period shows good concordance between hypothetical and actual anaesthetic techniques. Furthermore, we acknowledge that there are anaesthesia-related costs that were not measured, including fractional cost of non-disposables (for example, monitors), power usage, waste disposal, etc, and these might be expected to be higher for general anaesthesia due to greater use of disposable equipment. Equally, there is a failure rate for spinal anaesthesia of up to ∼ 4% [21], necessitating conversion to general anaesthesia, with a commensurate summation of costs. However, we estimate that the financial benefit of spinal compared with general anaesthesia is only negated above a spinal anaesthesia failure rate of 16%, or possibly during unexpectedly long operations that may require conversion to general anaesthesia [22].

Nationally, as ∼ 50% of hip fracture repairs are performed under general anaesthesia [11], we estimate that anaesthesia for hip fracture repair costs the UK ∼ £18 million annually. Yet in a questionnaire of anaesthetic intent for hip fracture repair, Sandby-Thomas et al. [23] found that ∼ 72% of 155 trauma anaesthetists preferred regional anaesthesia. If this proportion of regional anaesthetics were reflected nationally, we estimate an annual saving of ∼ £1.3 million.

Reduced government investment, rising costs and the prevailing financial climate are likely to intensify the pressure on clinicians to reduce costs over the next decade [24]. However, given that the relative cost of anaesthesia for patients undergoing hip fracture repair was found to be only 3.8% of the Payment by Results revenue received, it follows that even significant cost savings in anaesthesia are unlikely to have much financial impact. Nevertheless, providing that clinical quality is maintained, there remains some onus on anaesthetists to try and reduce costs. Traditionally, there has been pressure to use inexpensive drugs and to administer cheaper inhalational agents and/or use lower gas delivery rates [25]. Although the expense of some drugs (for example, ondansetron 4 mg at £5.39 was noted by the authors to be considerably higher than cheaper alternatives (for example, cyclizine 50 mg at £0.49, and the use of sevoflurane at high gas delivery flow rates (particularly during induction of anaesthesia) was notably more expensive than isoflurane at low gas delivery flow rates) has been highlighted drugs and inhalational agents only contributed 12% of the cost of anaesthesia. Similarly, the cost of simple airway equipment represents a relatively insignificant target for potential cost savings.

We noted three items of relatively high expense, namely single-use forced air warming blankets (£17.82 unit−1), fluid warmers (£18.78 unit−1), and suction bottle liners (£63.45 unit−1). Although the use of warming devices is supported by the National Institute of Health and Clinical Excellence [26], cost analysis would be a valuable addition to any such future research. Avoidance of tracheal intubation may reduce the cost of general anaesthesia, by reducing the usage of suction equipment (in addition to the additional costs of neuromuscular blocking and reversal agents).

Personnel costs could be theoretically reduced by utilising junior anaesthetic staff or anaesthetic practitioners, but these may be counterproductive measures and should be balanced against the cost ramifications of possible increases in the length of anaesthesia [25], potential for peri-operative complications and increased rates of operative deferment.

Operative costs have been calculated previously to account for only 9% (£1 095) of the estimated £12 163 mean hospital expenditure per hip fracture [8]. Our results indicate that anaesthesia contributes approximately only a fifth of operative costs. Indeed, the total cost of anaesthesia for cemented hemi-arthroplasty (Thompson’s) is similar to that of the prosthesis inserted, before accounting for the costs of operating room personnel and equipment. Furthermore, ward costs account for 84% of expenditure, and at a daily rate of approximately £500, this suggests that measures aimed at reducing the length of hospital stay (for example, through early operation [27], use of multidisciplinary integrated care pathways [28] and orthogeriatric input [29]) are likely to produce the greatest financial savings. We would also suggest that there might be a financial incentive in amalgamating hospital and social care budgets for patients sustaining hip fracture, to expedite early transfer to community rehabilitation facilities. It is worth highlighting that there is considerable disparity, £2090–£6253, between the estimated mean hospital expenditure, £8000–£12 163 [8, 9], and the mean tariff received by our Trust through Payment by Results of £5910 (which is similar to the national 2007 mean tariff of £5305 across all Trusts) [30]). In other words, many hospitals are effectively losing money when treating patients with hip fracture. Whilst this should properly stimulate a search for potential cost savings, it might also suggest that a recalculation of the tariffs (or of Payment by Results as a whole) is warranted.

In the absence of convincing differences in clinical outcomes between spinal and general anaesthesia [14], we propose that cost should be one reasonable factor determining the ultimate choice. Our results suggest that spinal anaesthesia offers a saving of ∼ £80 per case compared with general anaesthesia for hip fracture surgery and individual anaesthetists might take this account when planning an anaesthetic for their patients.

Acknowledgements and conflicts of interest

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements and conflicts of interest
  7. References

The authors would like to thank their anaesthetic and operating theatre colleagues for their support during the conduct of the survey. SW is currently an expert adviser involved in the formulation of National Institute for Health and Clinical Excellence guidelines on hip fracture management. No external funding and no other competing interests declared.

References

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements and conflicts of interest
  7. References