Intranasal ketorolac in renal colic

Renal colic is painful. Very, very painful. Patients consistently rate it as the one of the most painful conditions one can experience – worse than dislocations, broken bones¹, and even childbirth². It is usually described as a “stabbing” sensation between the flank and groin, accompanied by nausea and vomiting. There is sometimes – but not always! – blood in the urine.

Diagnosing renal colic is rarely difficult. Most of these patients can be picked up at triage³. The challenge for emergency clinicians is getting the pain under control. Oral medications are a no-go: they take too long to work, and patients with renal colic tend to vomit them back up. Guidelines recommend parenteral non-steroidal anti-inflammatories (NSAIDs) due to their hypothesised action on the ureteric smooth muscle.

Here in Virchester, we use ketorolac, which is one of several NSAIDs licensed in the UK. There is good evidence supporting ketorolac⁴ and I have personally seen it work wonders in renal colic. However, we tend to administer it intravenously, which creates delays. Patients need to be cannulated, doses second-checked, etc. Ketorolac is not currently something a triage nurse can prescribe and give.

A study recently published in the Annals of Emergency Medicine⁵ investigated whether intranasal ketorolac is an effective alternative for pain relief in renal colic. Specifically, the researchers wanted to know whether giving these patients ketorolac via the intranasal route is as effective as giving it intravenously.  

The abstract is below – followed by a summary of the paper – but as ever, I recommend that you read the full text and come to your own conclusions.

Abstract

Study objective: Atomized intranasal (IN) drug administration offers an alternative to the intravenous (IV) route. We aimed to evaluate the analgesic efficacy of IN versus IV ketorolac in emergency department patients with acute renal colic.
Methods: We conducted a double-blind, randomized controlled trial on adult patients (aged 18 to 64 years) with severe renal colic and numerical rating scale pain ratings ≥7.0. They were randomly assigned (1:1) to receive single doses of either IN or IV ketorolac. Our main outcomes were differences in numerical rating scale reduction at 30 and 60 minutes. A 95% confidence interval (CI) was calculated for each mean difference, with a minimum clinically important difference set at 1.3 points. Secondary outcomes included treatment response, adverse events, rescue medications, and emergency department revisits. We analyzed using intention-to-treat.
Results: A total of 86 and 85 patients with similar baseline characteristics were allocated to the IV and IN groups, respectively. Mean numerical rating scale scores were 8.52 and 8.65 at baseline, 3.85 and 4.67 at 30 minutes, and 2.80 and 3.04 at 90 minutes, respectively. The mean numerical rating scale reduction differences between the IV and IN groups were 0.69 (95% CI -0.08 to 1.48) at 30 minutes and 0.10 (95% CI -0.85 to 1.04) at 60 minutes. There were no differences in secondary outcomes.
Conclusion: Neither IN or IV ketorolac was superior to the other for the treatment of acute renal colic, and both provided clinically meaningful reductions in pain scores at 30 to 60 minutes.

What type of study is this?

This is a randomised controlled trial, which is the right methodology for this clinical question.

An observational study, such as a chart review or cohort study, would have been limited by selection bias. The participants in observational studies are – broadly speaking – a skewed sample. They are not prescribed the research drug by random chance, but because their doctors believe they should have it. These beliefs can exert a powerful influence on results.

For example, take an assumption commonly held by clinicians: that medications “work better” if given intravenously⁶. Had this trial been an observational study, participating doctors might have reserved intravenous ketorolac for the patients they felt needed a “better” treatment – i.e. those with more severe pain. This could distort results in favour of intravenous ketorolac, as patients with a worse baseline are generally more likely to improve due to an effect called ‘regression to the mean’. By randomly allocating patients to treatment, the authors of this trial ensured a fair distribution of acuity in the intervention and control groups, reducing an important source of bias in their results.

Where and when was this study done?

This trial was undertaken in the emergency department of a tertiary academic hospital in Oman between 2020 and 2022. It is unclear if patients were recruited out-of-hours.

Tell me about the patients

Patients were considered for this trial if they had severe flank pain and were suspected to have renal colic on the basis of their history and examination.

Exclusion criteria included contraindication(s) to ketorolac, recent NSAID use, and haemodynamic instability. In theory, pregnant or breastfeeding women could not participate, but it would appear that none were approached.

In total, 171 patients were included in this study. The majority (83%) were male, and the mean age was 35. Roughly two-thirds had the diagnosis of renal colic confirmed on subsequent imaging. The average stone size was five millimetres.

Baseline pain was self-reported on a numerical rating scale. To participate in the study, the pain had to be “severe” – 7/10 or worse – but the mean pain score before ketorolac was 8.5/10.

How were they treated?

Patients were randomised to one of two treatment groups. They were not aware of which group they were allocated to, and neither were their treating clinicians. All medications were prepared by a nurse who was not involved in the patient assessment process, and labels were not used

The intervention group (n = 85) received 30mg of intranasal ketorolac (0.5ml in each nostril) in addition to an intravenous placebo solution consisting of 10ml of 0.9% saline.

The control group (n = 86) were given the opposite: 30mg of intravenous ketorolac (3mg / ml) and an intranasal placebo solution composed of 1ml of 0.9% saline (0.5ml in each nostril).

What were the main outcome measures?

The primary trial outcome was a change in self-reported pain scores at thirty and sixty minutes post-treatment.

This outcome was recorded on an intention-to-treat basis – meaning that data were analysed according to the treatment patients were randomised to, even if they did not receive it or “crossed over” to another treatment. There are numerous advantages to this approach, which have been discussed on St Emlyn’s previously.

A number of secondary outcomes were used, including the prescription of rescue medications and the occurrence of adverse events, such as allergic reactions.

Confidence intervals were drawn around group means to establish the statistical significance of any differences between groups at baseline, thirty minutes, and sixty minutes post-treatment. There was mention by the authors of Student’s t-test, but I could not see evidence of its use in the published manuscript.

What did they find?

At thirty minutes, pain scores had fallen by an average of 4.67 from baseline in the intravenous ketorolac group and 3.98 in the intranasal group. The 95% confidence interval drawn around this difference was between -0.08 and 1.48, indicating statistical non-significance.

At sixty minutes, pain scores had fallen by an average of 5.71 from baseline in the intravenous ketorolac group and 5.61 in the intranasal group. Again, the difference between groups was not statistically significant.

There were no significant differences between groups on any secondary outcomes, and no serious adverse effects of ketorolac were observed. The most commonly reported side-effects of ketorolac reported were nausea and vomiting, although I would expect to see this in patients with renal colic anyway.

The bottom line: Ketorolac gave these patients impressive pain relief, and giving it intra-nasally did not appear to make it any less effective.

What should we take away from this?

This trial supports ongoing efforts to improve the management of acute pain in emergency care. RCEM describe this as an ‘under-recognised, under-treated, and under-prioritised’ problem in UK hospitals. I have to agree. Too many patients come to us in crisis and get fobbed off with co-codamol. Trials like this show that there is much more we can do for them. I can imagine intranasal ketorolac being prescribed and administered in triage.

With that said, there are some caveats to keep in mind as we interpret this paper. These were discussed at our journal club earlier this year and felt to be the most significant points for critical appraisal.

Firstly, this should really have been a non-inferiority trial. It is highly unlikely that the analgesic effect of intranasal ketorolac could be superior to intravenous ketorolac. The key question is whether the intranasal preparation is close enough (i.e. “not unacceptably worse”) for us to start thinking about its other advantages. Had the trial demonstrated non-inferiority of intranasal ketorolac but quicker pain control and reduced nursing burden, we would have more of a case for introducing it here in Virchester.

Of course, including these additional outcomes would have created other problems. The patients and clinicians in this trial were masked to treatment allocation because intranasal and intravenous compounds were given in all cases. Hence, the nursing burden and time-to-analgesia were, by design, identical between groups. It is difficult to imagine how masking could be maintained and these outcomes considered.

Another limitation to this study is its exclusion criteria. The most common reason for non-participation (n = 12) was prior use of another NSAID. We felt that this was difficult to justify in an otherwise pragmatic sample. These are, after all, patients in acute pain. Many will use every medication available to them before coming to the emergency department, and this often includes ibuprofen. Is ketorolac as effective for these patients? A sub-group analysis would be revealing here.

Finally, the authors should be commended for using a patient-centred outcome measure, but we should keep in mind that there are other ways to measure pain in clinical trials. Many patients are stoic – or wish to appear so – and under-rate their pain when asked about it by doctors. With these patients, objective findings such as tachycardia, sweating, and grimacing can additionally be used to guide pain relief. In clinical practice, I often use a combination of the patient’s self-reported pain and their observable behaviour in the waiting room to guide my prescribing. I would personally be very interested to see a composite pain outcome used in a trial like this one.

Should this study change our practice?

Not yet. This trial suggests a viable alternative to intravenous medication for patients with renal colic, but as one of the attendees at our journal club rightly observed, there is already a widely-used alternative with some support in the literature, and this is rectal diclofenac⁷. I would be curious to see how intranasal ketorolac compares to this unofficial “standard of care” in many UK emergency departments. 

It is worth pointing out that there is currently no licensed preparation of ketorolac for nasal administration in the UK, so it is unlikely we will be trying it here in Virchester any time soon.

References

1. Daoust R, Paquet J, Piette É, Sanogo K, Bailey B, Chauny JM. Impact of age on pain perception for typical painful diagnoses in the emergency department. The Journal of Emergency Medicine. 2016 Jan 1;50(1):14-20. ↩︎
2. Miah S, Gunner C, Clayton L, Venugopal S, Boucher NR, Parys B. Renal colic and childbirth pain: female experience versus male perception. Journal of Pain Research. 2017 Jul 5:1553-4. ↩︎
3. Loo CJ, Pouryahya P. Clinical Accuracy of Renal Colic Diagnosis before Imaging; Are we Biased?. Archives of Internal Medicine Research. 2020;3(4):235-41. ↩︎
4. Alghamdi YA, Morya RE, Bahathiq DM, Bokhari AF, Alaboud AK, Abdulhamid AS, Ghaddaf AA, Jamjoom M. Comparison of acetaminophen, ketamine, or ketorolac versus morphine in the treatment of acute renal colic: A network meta-analysis. The American Journal of Emergency Medicine. 2023 Nov 1;73:187-96. ↩︎
5. Al-Khalasi US, Al AK, Al-Jufaili M, Al-Reesi A, Al-Zakwani I, Al-Asmi MS, Al-Riyami FB, Vishwakarma R. Atomized Intranasal Ketorolac Versus Intravenous Ketorolac for the Treatment of Severe Renal Colic in the Emergency Department: A Double-Blind, Randomized Controlled Trial. Annals of Emergency Medicine. 2024 Mar 1;83(3):217-24. ↩︎
6. Broom J, Broom A, Adams K, Plage S. What prevents the intravenous to oral antibiotic switch? A qualitative study of hospital doctors’ accounts of what influences their clinical practice. Journal of Antimicrobial Chemotherapy. 2016 Aug 1;71(8):2295-9. ↩︎
7. Hosseini MM, Yousefi A, Ghahramani L, Rastegari M, Ebrahimi AR. Comparison of the Therapeutic Effects of Rectal Diclofenac Sodium and Intramuscular Pethidine Injection in the Treatment of Acute Renal Colic: A Randomized Clinical Trial. J Clin Trials. 2015;5(225):2167-0870. ↩︎