Medical practice

Hip replacement deaths drop by a half since 2003

“Death rates following hip replacement surgery fell by half in England and Wales,” reports the BBC News website.

Its headline is based on a new study in The Lancet which looked at data from the National Joint Registry (NJR) over the course of eight years. The registry is an NHS database recording outcomes in artificial joint operations such as hip and knee replacements.

Historically, the risk of deaths occurring during the first 90 days following a hip replacement – often due to complications such as blood clots has been relatively low. In 2003, NJR figures put post-surgical mortality at around one in 180 (0.56%). However, a single death as a result of treatment is still one too many, and so the researchers wanted to find out what factors could be linked to increased or decreased risk so they could identify ways to possibly reduce the risk further.

Encouragingly there has been a steady year by year fall in mortality, from 0.56% in 2003 to 0.29% in 2011 – roughly halving the death rate.

The researchers found that factors associated with decreased risk were a surgical approach that went in through the muscles in the buttocks (gluteal muscles), use of spinal (local) rather than a general anaesthetic, and use of compression stockings and medications to reduce risk of blood clots. 

The researchers acknowledge the possibility that other factors may be influencing the results such as patient health and the presence of other medical illness.

Nevertheless, the findings show a promising picture, suggesting that clinical guidelines about best practice, such as guidance about preventing blood clots (PDF, 5.8MB), can save lives.

Where did the story come from?

The study was carried out by researchers from the University of Bristol, University of Exeter, University of Oxford and Norwich Medical School.

The study was published in the peer-reviewed medical journal, the Lancet, and funded by the National Joint Registry for England and Wales.

BBC News reporting of the study was both useful and accurate.

What kind of research was this?

This study used data from national databases for England and Wales to examine trends in death rates following total hip replacement for osteoarthritis (so-called “wear and tear arthritis”) over the eight year period from April 2003 to December 2011. The researchers also looked at which factors related to treatment and patients were associated with mortality.

Total hip replacement (THR) is a major operation and the researchers say that although death following THR is rare, the actual size of the risk needs to be quantified. By doing this the researchers hoped to see which factors are associated with death following surgery, and subsequently what healthcare teams might be able to do to reduce risk to patients.

Current NHS guidance recommends measures to reduce the risk of a death after surgery, such as giving patients preventative treatments to reduce the risk of blood clots forming. However, the researchers say that as mortality after THR is low, it has been difficult to see which measures are most effective.

This research benefits from using a reliable dataset to inform mortality rates and which factors are associated with increased or decreased mortality. However, the data cannot tell us whether the factors identified are directly responsible for causing, or preventing, death.

What did the research involve?

The researchers gathered data about hip replacements performed between April 2003 and December 2011 from the National Joint Registry for England and Wales. Details of patients who had THR were then passed to the NHS Personal Demographics Service. They used patient NHS numbers to link to the Office for National Statistics to identify deaths from any cause within 90 days of the operation.

By linking to hospital inpatient and outpatient records the researchers also obtained sociodemographic details and information on other illnesses that the patients had in the five years prior to their THR.

The researchers used statistical methods to examine various factors associated with postoperative death (defined as death occurring up to 90 days after surgery) from any cause, including:

  • surgical technique
  • implant type and fixation method
  • anaesthetic type
  • preventative treatment for blood clots
  • age
  • sex
  • body mass index

Information on patients’ ethnicity, social deprivation and other medical illnesses was also analysed.

What were the basic results?

Over the eight year study period there were 458,568 THRs performed, and from those they excluded people whose THR was not performed for osteoarthritis, who had replacements of both hips performed at the same time, and for whom the NHS number could not be traced. This left 409,096 THRs included in the analyses, and of these, the average patient age was 68 years.

Sixteen percent of total hip replacements were privately funded, the funding source was uncertain for 6% and the remainder were NHS operations. Data on other medical illness and sociodemographics was available for 75% of patients. Just over half of the remaining quarter of cases for which no data was available were privately funded operations.

Of the 409,096 THRs, 1,743 patients died within 90 days of surgery during the eight year period (0.4% of all of those treated). Following adjustment for age, sex and other medical illnesses, there was a steady year by year fall in mortality over the eight year period, from 0.56% in 2003 to 0.29% in 2011 – roughly halving the mortality risk over the period.

The researchers identified several factors that were associated with significantly decreased mortality risk:

  • posterior surgical approach (through the gluteal muscles) 
  • treatment to prevent blood clots: both “mechanical” (such as with compression stockings) and medical (anti-clotting drugs such as heparin with or without aspirin) treatments independently reduced risk
  • use of spinal anaesthetic (numbing the area below the anaesthetic) rather than a general anaesthetic (making the person unconscious)

Surprisingly, being overweight was associated with lower mortality risk compared with being normal weight. However, the researchers warn that BMI data was missing for over half of participants, so these results should be interpreted with caution.

Unsurprisingly, severe liver disease, metastatic cancer, heart failure, history of heart attack and kidney disease were all associated with increased mortality risk.

How did the researchers interpret the results?

The researchers conclude: “Postoperative mortality after hip joint replacement has fallen substantially”. They suggest that adoption of management strategies of using a posterior surgical approach, spinal anaesthesia and mechanical and drug treatments to prevent blood clots, could reduce mortality risk further.

Conclusion

This is an informative study which demonstrates a decline in 90-day mortality rates following total hip replacement in England and Wales between 2003 and 2011.

The study benefits from using reliable datasets: the National Joint Registry for England and Wales is said to have recorded all total hip replacements performed since 2003; the Office for National Statistics records all deaths; and the Hospital Episode Statistics records data for all people who have received NHS-funded hospital care.

The research has identified factors associated with decreased or increased 90-day mortality, although it is not possible to say with certainty that any of these factors have directly caused or prevented postoperative death.

The research has attempted to identify and adjust for various other confounding factors that could be associated. But the researchers acknowledge that there is the possibility that some information on health and other medical illnesses has not been fully recorded. 

There is reportedly considerable debate over what is the best surgical approach to use for THR, and the choice is often influenced by conventional practice within different orthopaedic units, as well as patient characteristics.

The researchers suggest that the posterior surgical approach (through the muscles in the buttocks, rather than approaching from the side, or less commonly, from the front) may be associated with reduced risk because it is associated with less disruption to muscles, less bleeding and better mobilization after surgery.

In general, spinal anaesthesia is associated with less risk of complications than general anaesthetics, including better recovery, less need for pain relieving medications after surgery, and lower infection and bleeding risks. The researchers acknowledge that there may be patient related factors (such as fitter patients being more likely to undergo spinal anaesthesia) that may confound the results. They have tried to adjust for other medical illness, though as said, some information mayhave been missed.

Both drug and mechanical methods of preventing blood clots are widely recommended practice prior to many surgical procedures, so these factors being associated with reduced risk is perhaps unsurprising and confirms current recommendations.

The overall decline in mortality rates over time is encouraging, and may reflect the gradual general improvement in medical, surgical and anaesthetic practice, as well as the improved health of the aging population.


NHS Attribution