Wednesday, September 7, 2011

Hip Fracture: "A Quintessential Geriatric Illness"

Image from the National Library of Medicine nlm.nih.gov
I'd like to briefly turn readers attention to two terrific papers published recently in the Annals of Internal Medicine on the timing and outcomes of surgical hip fracture repair.

The first, an empirical study by Vidan and colleagues from Spain of 2250 patients with hip fracture, investigating the relationship between timing of hip fracture repair and the outcomes mortality and in-hospital complications (delirium, pneumonia, UTI, pressure sores).  Poorer outcomes have been observed for patients with delayed time to hip fracture repair compared to those with immediate hip fracture repair.  Surgeries are often delayed for patients who are very ill, such as patients with advanced heart failure, so that their physiology can be optimized prior to the stress of surgery.  The question is - are the worse outcomes due to the delay in surgery or the reason for the delay?

This study illustrates an interesting research principle called "selection bias."  In the raw unadjusted analysis, outcomes were worse for those with delayed time to hip fracture repair (less than 48 hours).  After adjustment for the severity of illness of patients prior to surgery, however, the outcomes were pretty much the same, regardless of the timing of surgery, except for patients with very long delays (greater than 5 days), where the outcomes were still worse.  Most of the differences in outcomes were due to severity of illness, not timing of surgery - younger healthier patients went to surgery earlier, and had better outcomes.  This selection bias explained most of the discrepancy in the relationship between timing of sugary and mortality/post operative complications.

In the accompanying editorial by William Hung (not that William Hung, a Geriatrician and palliative medicine doc at Mt. Sinai) and Sean Morrison (a Geriatrician at Mt. Sinai and immediate past president of the American Academy of Hospice and Palliative Medicine) described hip fracture as "the quintessential geriatric illness."  They say:
It is a rare patient with hip fracture who does not have a comorbid condition, such as dementia or frailty; multiple chronic diseases; or functional impairment, particularly with respect to locomotion or transfer, before the hip fracture event. Thus, medical reasons for delaying surgery are the rule rather than the exception. Although hip fracture is typically regarded as a surgical condition, the poor underlying health status of the patient population, coupled with the effects of hip fracture on functional status and existing comorbid conditions, means that nonsurgical aspects of care are critical to optimizing clinical outcomes.
Mortality and post-surgical complications aside, they argue for early hip fracture repair on the basis of evidence demonstrating improved pain control with early surgical intervention.

The big remaining question for me is not so much about the timing of surgery, but non-operative options for hip fracture.  Is not having surgery an option?  For which patients, and under what circumstances?  I briefly cared for a patient with advanced dementia who was mostly bed bound in the nursing home prior to falling and fracturing her hip.  The surgeons' recommended a non-surgical approach because: (1) she was non-ambulatory at baseline, (2) she would not be able to lie still after the repair or participate in rehabilitation, (3) she had a limited life-expectancy, (4) her fracture was non-displaced (the bones still lined up well), and (5) she was not in much pain, and her pain was responsive to simple analgesics (I think just tylenol after a few days).  I'm not sure how things turned out for her.  Have others had any experience with non-surgical management?  Does anyone know of any data?  In the absence of a surgical repair, I imagine pain control must be challenging.

by: Alex Smith

3 comments:

R. Vaughan, MD said...

There are studies which show some surprising results. In the non operative group death is more common due to pre-existing disease in the first month (not unexpected, this is why we didn't operate) but after a month the mortality, mobility, etc outcomes start to be the same regardless.
These are small studies and subject to flaw, but still impressive.

Russ Vaughan, MD

Michelle Allen, LMSW said...

My experience with non-surgical management after a hip fracture is purely anecdotal. I worked for years as a hospice social worker, and many times the hospice nurse and I would counsel the families of older, frail patients with dementia about treatment options after a hip fracture (or a fall with a potential hip fracture). Do you want to go to the hospital? Do you want her to have surgery? Do you want to keep her comfortable in bed and pursue no treatment? If the illness has been long, and if the person no longer recognizes their loved ones, the families will often opt for no surgery and even no trip to the ER. The hospice team has effectively managed the pain and kept the patients comfortable in bed (and most of these patients were residents of assisted living facilities), and death comes quickly and usually very peacefully. Like I said, though, only anecdotal.

Bear in the striped shirt said...

My anesthesia colleagues and I were having a conversation about this very topic just recently. Frequently we care for very sick, older patients with dementia who come to the OR for repair of hip fractures, and though we can get them through the operations, we hear from surgeons that their overall survival is poor.

It would benefit these older patients greatly to study the non-surgical management of hip fractures, as well as the post-operative course after surgery. So many patients and families think surgery is the final answer to any problem (not just fractures, but cancer, obesity) that if it is an option, it should be taken. They don't realize that the dangers of surgery don't end when the patient wakes up. The weeks after surgery are often when we see complications like MI, DVT, decubitus ulcers from immobilization, delirium, etc.

An interesting discussion.