Femoral nerve block for hip fractures: the evidence

The most recent topic that Rory Spiegel, Anton Helman and I covered for the Emergency Medicine Cases Journal Jam is femoral nerve blocks for hip fractures. This is my summary of the evidence.

Overall Summary

I thought the evidence for femoral nerve blocks would have been stronger than it is. We decided to cover this topic on the Journal Jam because we wanted an example of evidence based medicine with positive results (to balance some of the negatives, such as thrombolytics for stroke or BNP for anything). Ultimately, the discussion was more about how we should practice in the face of imperfect evidence, which is a more common scenario anyway. The evidence here is far from perfect. The studies were small. Most studies lacked blinding, which could introduce significant bias. Different techniques were used by different providers, resulting in significant heterogeneity. Furthermore, a lack of ultrasound makes extrapolating the results difficult, as ultrasound is probably now standardly used in emergency medicine. 

Although imperfect, I think these studies make it pretty clear that femoral nerve blocks will decrease pain and decrease opioid use. However, the decreases in pain are pretty modest, and a decrease in opioid use is really a surrogate outcome. At this point, we don’t have any evidence that the nerve block approach to pain management decreases adverse events from opioids, or decreases delirium. On the other hand, I think the evidence suggests that this is a fairly safe procedure.

After reviewing this literature, I will continue to offer a femoral nerve block to all of my patients with hip fractures. I think there is enough evidence here to suggest that the benefits will outweigh any harms, but I recognize that the evidence is imperfect, and am ready to adjust my practice as more studies are published.

 

Systematic review

For complex topics, I generally think that it is best to go straight to the original literature. That is why previous journal jams have involved lengthy discussions of multiple RCTs. When trying to understand the controversy that surrounds thrombolytics for stroke, or the confusion around contrast and kidney injury, a deep dive into the literature can really help. I don’t think that this is one of those topics. If you are trying to decide whether to start using femoral nerve blocks in your practice, you will probably find all the information you need in a systematic review. (I will briefly summarize the RCTs below, if you are interested.)

Ritcey B, Pageau P, Woo MY, Perry JJ. Regional Nerve Blocks For Hip and Femoral Neck Fractures in the Emergency Department: A Systematic Review. CJEM. 2016; 18(1):37-47. PMID: 26330019 [free full text]

Methods

• This is a systematic review with excellent methods and a good search.
• They included RCTs of adult patients with acute hip or femoral neck fractures who had a single nerve block done preoperatively.
• A femoral nerve block, 3 in 1 femoral nerve block, or fascia iliaca block were all acceptable.
• The primary outcome of interest was reduction in pain.

Results

• There were 9 RCTs that fit their inclusion criteria (2 looking at FNB, 4 at 3 in 1 block, and 3 FICB). Studies were small, ranging from 33 to 154 patients.
• The results were too heterogeneous to all for an appropriate meta-analysis, so the results had to be provided descriptively.
• Most included studies had a moderate to high risk of bias, with the biggest issue being a lack of blinding.
• The block was performed by an emergency physician in 5 of the 9 studies.
• Only one study used ultrasound guidance. 5 relied on landmarks and 3 used a nerve stimulator.
• Pain was better in the nerve block group in 8 of the 9 studies.
• Less opioids were used in the nerve block group in 8 of the 9 studies.
• No study reported any life threatening complications due to the nerve blocks, but the authors thought that the methodology was poor, and complications could have been under-reported.

Bottom line: Femoral nerve blocks decrease pain and probably decrease opioid use in patients with hip fractures. There is not enough data to determine if other outcomes such as delirium are impacted. These RCTs are not an adequate source for safety data.

RCTs

These are the major RCTs that look at this issue. In general, they are captured in the systemic review, and the results are pretty consistent. I include these as a reference, but most readers probably want to skip to the next section.

Haddad FS, Williams RL. Femoral nerve block in extracapsular femoral neck fractures. The Journal of bone and joint surgery. British volume. 1995; 77(6):922-3. PMID: 7593107

Brief summary: 50 consecutive emergency department patients with extracapsular fractures of the femoral neck were randomized to landmark based femoral nerve block using 0.3 ml/kg of 0.25% bupivacaine. Pain was better in the nerve block group at 15 minutes at 2 hours, but was no longer statistically so at 8 hours. Less IM opioid was used in the control group (35 vs 12 requests for a dose). There were statistically less respiratory complications in the nerve block group (2 vs 9).

Key caveats: No placebo or blinding; all blocks done by the same doctor (not sure how they actually got consecutive patients – does this doctor live in the hospital?); regular analgesia is not modern (pethidine and Voltarol); there is no stated primary outcome.

Fletcher AK, Rigby AS, Heyes FL. Three-in-one femoral nerve block as analgesia for fractured neck of femur in the emergency department: a randomized, controlled trial. Annals of emergency medicine. 2003; 41(2):227-33. PMID: 12548273

Brief summary: A RCT of 50 adult emergency department patients with femoral neck fractures, randomized to a landmark based 3 in 1 FNB with 20 mL of 0.5% bupivacaine or standard care with morphine. Patients with a nerve block had a faster time to their lowest recorded pain score (3 vs 6 hours), had lower pain score overall, and used less morphine (avg of 0.4 vs 1.2 mg/hr). There were no adverse events.

Key caveats: Excluded patients who were confused (dementia is very common in this population); patients and clinicians were not blinded, but the ward nurses recording pain were; clinical data was obtained by chart review; there is no stated primary outcome and they measured multiple outcomes.

Kullenberg B, Ysberg B, Heilman M, Resch S. [Femoral nerve block as pain relief in hip fracture. A good alternative in perioperative treatment proved by a prospective study]. Lakartidningen. 2004; 101(24):2104-7. PMID: 15282985

This article is in Swedish, so I can’t read it. They report lower pain scores, but if you look at the analysis in the Ritcey (2016) systematic review, although pain scores dropped from 6/10 to 2/10, they look very similar to the control group.

Foss NB, Kristensen BB, Bundgaard M, et al. Fascia iliaca compartment blockade for acute pain control in hip fracture patients: a randomized, placebo-controlled trial. Anesthesiology. 2007; 106(4):773-8. PMID: 17413915

Brief summary: A RCT including 48 adult emergency department patients with clinical signs of hip fracture, randomized before x-ray to get either a landmark based FICB with 40 mL of 1% mepivacaine or placebo (a saline injection). It took 4 minutes to place the block. There was no difference between the groups in terms of pain after treatment (2/10 in both groups at 60 minutes), but the groups were unbalanced at the start, with the FICB group starting with 5/10 pain and the morphine group starting with 2/10 pain. The nerve block group used less morphine (median 6 mg vs 0 mg, p<0.01).

Key caveats: This is one of the few blinded studies; they used a double dummy set up (both groups got an IM injection and an injection at the nerve block site); there was one patient who was randomized and turned out not to have a fracture; based on neurologic testing, only 67% of the nerve block group had evidence of a successful block.

Graham C, Baird K, McGuffie A. A Pilot Randomised Clinical Trial of 3-In-1 Femoral Nerve Block and Intravenous Morphine as Primary Analgesia for Patients Presenting to the Emergency Department with Fractured Hip Hong Kong Journal of Emergency Medicine. 2017; 15(4):205-211. [free full text]

Brief Summary: A RCT of 40 adult emergency department patients with femoral neck fractures, randomized to either a landmark based 3 in 1 FNB with 30 mL of 0.5% bupivacaine (not exceeding 3mg/kg) or morphine (0.1 mg/kg IV). It took 5-10 minutes to complete the nerve block. Pain scores were better in the nerve block group at 30 minutes, but not at 2, 6, or 12 hours. There was no difference in subsequent opioid use.

Key caveats: Not blinded; convenience sample; missing data for a bunch of patients; no clear primary outcome.

Monzón D, Vazquez J, Jauregui JR, Iserson KV. Pain treatment in post-traumatic hip fracture in the elderly: regional block vs. systemic non-steroidal analgesics. International journal of emergency medicine. 2010; 3(4):321-5. PMID: 21373300 [free full text]

Brief summary: A double blind RCT of 154 emergency department patients over the age of 65 with hip fractures, randomized to landmark based FICB with 0.3 mL/kg of 0.25% bupivacaine or an IV NSAID (diclofenac or ketorolac), with double dummies. At 15 minutes, pain was much better controlled in the nerve block group (scores 3/10 vs 6/10, p<0.001). The pain scores were about the same at 2 hours, and by 8 hours pain looked better controlled in the NSAID group, but not statistically so (2/10 vs 4/10, p=0.08). There were more adverse events in the NSAID group. There were 4 patients with delirium and 4 with nausea and vomiting, as compared to 0 in the nerve block group. The only complications with the nerve block group were local bruises at the injection site.

Key caveats: This is a well done, double blind, double dummy study; using NSAIDs rather than opioids in hip fractures is novel to me; no patients in the study received opioids; allocation concealment was inadequate; the treating nurse was unblinded.

Beaudoin FL, Haran JP, Liebmann O. A comparison of ultrasound-guided three-in-one femoral nerve block versus parenteral opioids alone for analgesia in emergency department patients with hip fractures: a randomized controlled trial. Academic emergency medicine. 2013; 20(6):584-91. PMID: 23758305 [free full text]

Brief summary: A blinded RCT of 38 emergency department patients over 55 years old with femoral neck or intertrochanteric fractures, randomized to ultrasound guided FNB with 25 mL of 0.5% bupivacaine plus morphine or morphine alone (plus sham nerve block). Pain scores were better in the nerve block group at all time points (15 minutes, 1, 2, and 4 hours). Pain decreased from 8/10 to 4/10). The scores never improved in the morphine group. Patients in the nerve block group used less IV opioids (median 0 vs 5mg, 0=0.28). There were no statistically significant differences in adverse events, but the numbers were lower in the nerve block group.

Key caveats: This study probably has the best methods overall; they clearly under treated with opioids, considering that the median given in the control group was only 5 mg, and the pain scores never changed in that group (this may be a fair representation of some real life practice); like all these trials, this is significantly underpowered for harms.

Fujihara Y, Fukunishi S, Nishio S, Miura J, Koyanagi S, Yoshiya S. Fascia iliaca compartment block: its efficacy in pain control for patients with proximal femoral fracture. Journal of orthopaedic science. 2013; 18(5):793-7. PMID: 23744530

Brief summary: This is a RCT of 56 adult patients with proximal femur fractures, randomized to either a landmark based FICB or NSAIDs (diclofenac 25mg rectally). The block was also repeated post-operatively. The nerve block worked great, with pain scores dropping from 9/10 to 3/10 by 10 minutes, and staying there 12 hours later. Not surprisingly, the NSAID had no real effect on the pain.

Key caveats: It seems cruel to treat hip fractures with only a rectal NSAID; there was no allocation concealment, and no blinding.

McRae PJ, Bendall JC, Madigan V, Middleton PM. Paramedic-performed Fascia Iliaca Compartment Block for Femoral Fractures: A Controlled Trial. The Journal of Emergency Medicine. 2015; 48(5):581-589.

Brief summary: In this study, nerve blocks were done by paramedics. A RCT of adults with hip or femur fractures and a pain score over 5/10, randomized to FICB (anatomical landmarks) or standard care. All patients had 0.1mg/kg of morphine IV prior to the block. Pain reduction at 15 minutes was better in the nerve block group (reduction of 5/10 vs 2/10, p=0.025). Pain also looks better in the nerve block group on arrival to hospital, on transfer to the ED bed, and at 120 minutes, but only transfer to the ED bed was statistically significant. Intravenous morphine use was less in the nerve block group. There were no adverse events in the nerve block group, but 2 (both thought to be medication side effects) in the standard care group.

Key caveats: Done by paramedics; includes femoral shaft fractures; used lidocaine instead of a long acting agent; no placebo; one patient couldn’t get the block because landmarks were difficult.

Unneby A, Svensson O, Gustafson Y, Olofsson B. Femoral nerve block in a representative sample of elderly people with hip fracture: A randomised controlled trial. Injury. 2017; 48(7):1542-1549. PMID: 28501287 [free full text]

Brief summary: A RCT of 266 patients over 70 years old on an orthopedic ward with hip fractures, randomized to FNB with 40ml of 0.25% levobupivacaine by anesthesiologists using a nerve stimulator or standard care. Pain scores were statistically better after the nerve block, but the difference was small. Fewer patients in the nerve block group required opioids (40% vs 83%). There were no adverse events.

Key Caveats: The study didn’t take place in the emergency department; pain scores were only measured when the patient was at rest and were very low; by including patients with significant dementia, measurement of pain scores can be difficult

What is the best method?

There are 3 major techniques for providing peripheral nerve blockade in the setting of a femoral fracture: a femoral nerve block, a “3 in 1” femoral nerve block, and a fascia iliaca block. There are a couple studies comparing the efficacy of these different techniques:

Newman B, McCarthy L, Thomas PW, May P, Layzell M, Horn K. A comparison of pre-operative nerve stimulator-guided femoral nerve block and fascia iliaca compartment block in patients with a femoral neck fracture. Anaesthesia. 2013; 68(9):899-903. PMID: 23789738

Brief summary: RCT of 110 adult patients with femoral neck fractures comparing FICB to a nerve stimulator guided FNB. Reduction in pain was better the FNB block group, but this might be below a clinically detectable difference (a reduction of 2.8 vs 3.7/10, p=0.47). Less morphine was used in the FNB group.

Reavley P, Montgomery AA, Smith JE, et al. Randomised trial of the fascia iliaca block versus the ‘3-in-1’ block for femoral neck fractures in the emergency department. Emerg Med J. 2015; 32(9):685-689. PMID: [free full text]

Brief summary: RCT of 162 adult patients with femoral neck fractures, randomized to FICB or 3 in 1 block. Pain scores were the same at 30 and 60 minutes. Analgesic use between the two groups was the same. Pain scores were the same whether the block was done using ultrasound, a nerve stimulator, or anatomical landmarks.

My take: There isn’t a lot of evidence to go on. Any technique is probably fine. The specific method probably matters more when performing these blocks without ultrasound guidance. My technique is probably a hybrid. I try to be near the femoral nerve, but ensure I leave some space for safety. Then, I just make sure I am under the fasicia iliaca and watch the anesthetic spread around the nerve. If I have an extra set of hands, I will apply some distal pressure as in the 3 in 1 block, but usually I am doing this by myself.

Harms

The is no evidence of harm in the RCTs, but none of the RCTs were powered for or properly designed to detect harms.

When looking for harm data, I tend to search observational literature. The numbers in observational trials are large enough that rare harm events can be seen. However, because the data is observational, one cannot be sure about cause and effect. This is especially true when looking at this data, as patients receiving femoral nerve blocks for hip fractures go on to have major surgery, often accompanied by either spinal or general anesthesia. In that context, it is very difficult to know which intervention was the cause of the adverse event, assuming it wasn’t a consequence of the original trauma. This literature is also complicated by varying techniques, including the preponderance of blind (non ultrasound guided) blocks and the use of continuous infusion catheters in anesthesia. There are not enough emergency medicine specific studies, so I have had to extrapolate from the anesthesia literature.

Overall, the chance of a significant adverse event following a one-time femoral nerve block appears to be quite low.

Local anesthetic systemic toxicity (LAST) is very rare, with a rate between 8 and 30 in 100,000 patients, (Sites 2012; Barrington 2009; Orebaugh 2009) and can probably be limited by good technique and the use of ultrasound. That being said, I think anyone performing these blocks should know how to manage local anesthetic systemic toxicity.

There may be a risk of neuropathy, nerve damage, or persistent neurologic symptoms after a peripheral nerve block. One large systematic review reports a rate of 0.3% for femoral nerve blocks, and 3% for all peripheral nerve blocks. (Brull 2007). Other studies report rates between 2/10,000 and 4%, with the vast majority of these symptoms resolving on their own over the following months. (Auroy 1997; Fredrickson 2009; Sharma 2010; Sites 2012) However, as was mentioned above, neurologic symptoms are common after sugery, with reported rates that looks very similar to what is reported here. (Jacob 2011) In fact, in the 1 study that mentions a control group that underwent surgery, but didn’t have a peripheral nerve block placed, there was no difference in the rate of postoperative neuropathy. (Sharma 2010) Therefore, it is possible that peripheral nerve blocks do no contribute to neuropathies, but at worst this is a rare, generally minor, and self resolving adverse event.

There are a number of minor side effects to be aware of, including arterial puncture (2/1,000), venous puncture (1/1,000), and minor infections. (Sites 2012) There have been reports of rare retroperitoneal hematomas and psoas abscesses, but in the context of continuous catheter use rather than single injections. (Wigel 2007; Adam 2003) Falls is another potential risk, addressed in the next section.

I have heard of individual cases of hypotension, with the common feature being larger doses of opioids being given prior to the block. (Hypothesis: the block removes the painful stimulus, and the opioid side effects are unmasked?) However, hypotension was not reported as a side effect in any of the studies I reviewed.

Should I perform the block before the x-ray? (Fall risk?)

I have discussed this with a number of experts, and the general consensus is that hip fracture is a clinical diagnosis, and in the context of an obvious fracture, most people are comfortable performing the nerve block before x-ray. Some add ultrasound to help confirm the diagnosis.

It makes a lot of sense to me to place the block before x-ray, because x-ray will probably be the most painful part of our patients’ emergency department stays. However, this has not been my practice to date. I have been worried about a misdiagnosis: what if I place a block in a patient who doesn’t have a fracture? (This did occur in Foss 2007). Could the block prevent the patient from being discharged home? The femoral nerve innervates the extensors of the knee (quadriceps femoris and sartorius) in addition to its sensory innervation. A full femoral nerve block will result in weakness to knee extension, which could impact walking and the ability to safely discharge elderly patients home from the emergency department (especially as I want to use long acting agents to control pain).

Falls are reported as a potential adverse events after femoral nerve block in multiple studies. (Kandasami 2009; Sharma 2010) However, these are surgical patients receiving multiple medications, so falls might be expected. In the one study that comments on a control group, the fall rate was the same whether or not patients received a femoral nerve block. (Sharma 2010)

There is evidence that femoral nerve blocks result in weakness. In one study that used knee buckling while standing to assess for significant quadriceps weakness, it was seen in 29% of the femoral nerve block group as compared to 3% of controls (no statistics presented) on postoperative day 1. (Yadeau 2005) In a study of healthy volunteers, a significant decrease in knee extension strength was seen (11% of baseline) at 30 minutes, although it seemed to start to recover at 60 minutes (42% of baseline). (Kwofie 2013) In this study, they were using a shorter acting agent (chlorprocaine). In the same study, walking was impaired (based on a validated score) at 30 minutes, but the score was just below the cutoff for impairment.

In summary, we should expect some weakness in knee extension after a femoral nerve block, but it isn’t clear how long this weakness will last after a single injection in the emergency department, nor how clinically important it is. I will probably start performing the nerve block before x-ray when I think there is an obvious fracture if the patient wants it performed.

Discussion

There are a number of problems with these studies, leaving a lot of uncertainty about their results.

One major problem was a lack of blinding. Only 2 of these studies were appropriately blinded, which is a big problem when measuring inherently subjective outcomes like pain. The studies using a sham block had similar outcomes to the unblinded studies, which is reassuring, but if you studying this topic in the future, you really should include a sham procedure and completely blind providers, patients, and outcome assessors.

I am always concerned that harm may be under-reported in RCTs. There are a number of reasons that this can happen. Trials generally aren’t powered for rare harm events. It would only take 1 bad outcome from local anesthetic systemic toxicity (LAST) to outweigh any benefit seen here, but the trials are too small to see such a rare event. And there are some other reasons trials miss harms. They often pick healthier patients, or they use a run-in period to exclude patients with side effects, or they just may not look (it’s expensive and difficult to follow everyone for a long time). That being said, the best evidence I can find is that these blocks are very safe.

The heterogeneity across the studies makes them difficult to assess as a group. Some blocks were done by trainees, some by emergency physicians, and some by anesthesiologists. There are 3 different general techniques used, and within those techniques there are a large number of variations. Most importantly, only one study used ultrasound as part of the procedure, which I think is likely to drastically increase the accuracy and safety of this procedure.

Any study of hip fractures will run into the dilemma of what to do with patients with dementia. Including patients with advanced dementia makes use of standard pain scales difficult or inconsistent. On the other hand, excluding such patients would exclude a large proportion of the patients we would want to perform this procedure on.

I will note that in the studies that reported it, this is a very quick procedure to perform: between 5 and 10 minutes depending on the study.

 

Conclusion

The evidence is imperfect, but I currently offer a femoral nerve block to all my patients with a hip fracture, and I will continue that practice after reading these studies.

 

Other FOAMed

5 min sono fascia iliaca block video

Femoral nerve block on the ultrasound podcast

Ultrasound guided femoral nerve block on EMDocs

Other references

Amin NH, West JA, Farmer T, Basmajian HG. Nerve Blocks in the Geriatric Patient With Hip Fracture: A Review of the Current Literature and Relevant Neuroanatomy. Geriatric orthopaedic surgery & rehabilitation. 2017; 8(4):268-275. [pubmed]

Auroy Y, Narchi P, Messiah A, Litt L, Rouvier B, Samii K. Serious complications related to regional anesthesia: results of a prospective survey in France. Anesthesiology. 1997; 87(3):479-86. [pubmed]

Barrington MJ, Watts SA, Gledhill SR, et al. Preliminary results of the Australasian Regional Anaesthesia Collaboration: a prospective audit of more than 7000 peripheral nerve and plexus blocks for neurologic and other complications. Regional anesthesia and pain medicine. ; 34(6):534-41. [pubmed]

Brull R, McCartney CJ, Chan VW, El-Beheiry H. Neurological complications after regional anesthesia: contemporary estimates of risk. Anesthesia and analgesia. 2007; 104(4):965-74. [pubmed]

Fredrickson MJ, Kilfoyle DH. Neurological complication analysis of 1000 ultrasound guided peripheral nerve blocks for elective orthopaedic surgery: a prospective study. Anaesthesia. 2009; 64(8):836-44. [pubmed]

Jacob AK, Mantilla CB, Sviggum HP, Schroeder DR, Pagnano MW, Hebl JR. Perioperative nerve injury after total hip arthroplasty: regional anesthesia risk during a 20-year cohort study. Anesthesiology. 2011; 115(6):1172-8. [pubmed]

Kandasami M, Kinninmonth AW, Sarungi M, Baines J, Scott NB. Femoral nerve block for total knee replacement – a word of caution. The Knee. 2009; 16(2):98-100. [pubmed]

Kwofie MK, Shastri UD, Gadsden JC, et al. The effects of ultrasound-guided adductor canal block versus femoral nerve block on quadriceps strength and fall risk: a blinded, randomized trial of volunteers. Regional anesthesia and pain medicine. 2013;38(4):321-5. [pubmed]

Morrison RS, Magaziner J, Gilbert M, et al. Relationship between pain and opioid analgesics on the development of delirium following hip fracture. The journals of gerontology. Series A, Biological sciences and medical sciences. 2003; 58(1):76-81. [pubmed]

Orebaugh SL, Williams BA, Vallejo M, Kentor ML. Adverse outcomes associated with stimulator-based peripheral nerve blocks with versus without ultrasound visualization. Regional anesthesia and pain medicine. ; 34(3):251-5. [pubmed]

Sieber FE, Mears S, Lee H, Gottschalk A. Postoperative opioid consumption and its relationship to cognitive function in older adults with hip fracture. Journal of the American Geriatrics Society. 2011; 59(12):2256-62. [pubmed]

Sharma S, Iorio R, Specht LM, Davies-Lepie S, Healy WL. Complications of femoral nerve block for total knee arthroplasty. Clinical orthopaedics and related research. 2010; 468(1):135-40. [pubmed]

Sites BD, Taenzer AH, Herrick MD, et al. Incidence of local anesthetic systemic toxicity and postoperative neurologic symptoms associated with 12,668 ultrasound-guided nerve blocks: an analysis from a prospective clinical registry. Regional anesthesia and pain medicine. 2012;37(5):478-82. PMID: 22705953

Wiegel M, Gottschaldt U, Hennebach R, Hirschberg T, Reske A. Complications and adverse effects associated with continuous peripheral nerve blocks in orthopedic patients. Anesthesia and analgesia. 2007; 104(6):1578-82, table of contents. [pubmed]

YaDeau JT, Cahill JB, Zawadsky MW, et al. The effects of femoral nerve blockade in conjunction with epidural analgesia after total knee arthroplasty. Anesthesia and analgesia. 2005; 101(3):891-5, table of contents. [pubmed]