American Journal of Emergency Medicine 47 (2021) 13-16

Contents lists available at ScienceDirect

American Journal of Emergency Medicine

journal homepage:

Imaging in suspected ureteral colic: Creating new decision rules based on multispecialty consensus

Edward J. Durant ^a,b,?, David R. Vinson ^c,d

^a Kaiser Permanente Modesto Medical Center, 4601 Dale Road, Modesto, CA 94536, USA

^b Kaiser Permanente Bernard J. Tyson School of Medicine, USA

^c Kaiser Permanente Northern California Division of Research, Oakland, CA, USA

^d Kaiser Permanente Roseville Medical Center, Roseville, CA, USA

1. Introduction

Acute flank pain is a common complaint in the emergency depart- ment (ED), accounting for over 2 million visits annually [1]. Computed tomography (CT) is considered the gold-standard for diagnostic imag- ing in suspected ureteral colic, and it is performed in more than 90% of patients diagnosed with Ureteral stone [1,2]. The widespread use of CT is not without harm, particularly with regard to increased lifetime can- cer risk as a result of exposure to ionizing radiation [3,4]. Advanced di- agnostic imaging is also a major contributor to the rising costs of healthcare. The use of CT may not even be necessary for most of these patients. Only 10% of patients who present to the ED with acute flank pain are admitted for a Clinically important stone or non-stone diagnosis [2,5-7]. Developing a safe and simple clinical pathway to predict uncom- plicated ureteral stone without the use of CT has remained elusive.

Several researchers have attempted to develop clinical decision rules to predict uncomplicated ureteral stone without the use of CT. In 2014, Moore et al. published a retrospective derivation and prospective vali- dation of the STONE score, a clinical decision rule to predict the presence of uncomplicated ureteral stone and lower likelihood of alternative findings. The score is comprised of factors in the patient’s history, clini- cal characteristics, and laboratory findings [7]. The STONE score showed some initial promise in predicting the risk of ureteral stone and per- formed better than physician gestalt. On external validation, however, the high-likelihood category had a sensitivity of only 53% and specificity of 87%, which would not be sufficiently specific to defer a CT scan [6]. Additionally, the STONE score was not designed to exclude clinically im- portant ureteral stones (such as infected obstructing ureteral stone with sepsis) [6]. In 2017, Wang et al. derived two different high-sensitivity clinical decision rules to predict clinically-important stones and clinically-important alternative outcomes that required inpatient ad- mission [8]. Although these clinical decision rules provided excellent sensitivity, they had poor specificity and were not developed to predict uncomplicated ureteral stone.

More recently, Moore et al. published a literature review and multispecialty consensus to determine appropriate imaging for

* Corresopanding author at: Kaiser Permanente Modesto Medical Center, USA.

E-mail addresses: [email protected] (E.J. Durant), [email protected] (D.R. Vinson).

suspected ureteral colic, based on age, history of Kidney stones, and sev- eral other features of medical history, physical examination, Diagnostic findings, and response to treatment [9]. The authors composed 29 brief clinical vignettes with different permutations, including age, sex, pregnancy status, history of stones, and abdominal tenderness. All clin- ical vignettes were predicated on the presentation of acute flank pain associated with nausea, vomiting, and hematuria, with pain relief after intravenous analgesics. Possible imaging recommendations included ul- trasound (point-of-care or radiologist-performed), CT, reduced-dose CT, or no imaging. Consensus was sought from a 9-member panel, com- posed equally of urologists, radiologists, and emergency medicine phy- sicians. Agreement ranged from moderate to perfect. Consensus was defined a priori as perfect (9 of 9 agreement), excellent (8 of 9), good (6 or 7 of 9), moderate (5 of 9), or not reached. The group was able to reach excellent or perfect agreement in 23 of the 29 vignettes and good (7 of 9, in these cases) or moderate agreement in three cases each. The consensus document as written, however, is not readily trans- latable into clinical practice. Several intermediary steps were missing to move the consensus document toward clinical implementation [10]. We sought to create decision rules to begin to bridge the gap (Fig. 1).

1. Methods

We describe the creation of clinical decision rules that use the con- sensus recommendations as a foundation and scaffold. Although Moore et al., did not do so explicitly, the consensus recommendations could be represented as a clinical decision tree. The authors developed the 29 scenarios “to represent the best balance of major factors with the least number of scenarios…skewed toward those in which the clin- ical likelihood of a kidney stone is high.” We summarized most of the vi- gnettes as a clinical decision tree (Fig. 2) and addressed special cases falling outside of this decision tree separately (Fig. 3). It should be noted that the entry criteria for most of the clinical vignettes in the de- cision tree were presentation to the ED with acute onset of flank pain (<3 h), nausea or vomiting, hematuria on urinalysis or colorimetric point-of-care urinalysis, and pain relief with intravenous analgesics. Of the vignettes that included abdominal tenderness to palpation on phys- ical examination, only left lower quadrant tenderness was included. Presumably, the authors considered the risk of appendicitis in patients with right lower quadrant tenderness too high to include this finding

https://doi.org/10.1016/j.ajem.2021.03.041

0735-6757/(C) 2021

Fig. 1. The steps of translation from consensus document to clinical application.

Fig. 2. Summary of consensus based on clinical vignettes. LLQ TTP = left lower quadrant of the abdomen tenderness to palpation; POCUS = point-of-care ultrasound; RDCT = reduced dose computed tomography.

Fig. 3. Special cases in the evaluation of possible ureteral colic. ESWL = extracorporeal shockwave lithotripsy.

within the clinical pathway. The remaining four vignettes did not meet these entry criteria or did not fall neatly within the decision tree. Of these, two applied to patients who had recently undergone a urologic procedure and two applied to patients who already had a CT- diagnosed stone or intractable pain.

Once we had created our decision tree and identified special cases, we next translated the vignettes into simple and useful clinical decision

rules. We converted the singular ages in the original paper to age ranges for greater clinical utility. Age 12 years became “less than 18,” age 35 years became 18-44, age 55 years became 45-64, and age 75 years became “age 65 or greater.” This decision was necessarily arbitrary and the first of several ways in which our proposed clinical decision rules differed from vignettes for the sake of clinical utility, feasibility, or simplicity. Although consensus was good or better in all but three vi- gnettes, in the cases of moderate agreement, we opted for the more con- servative imaging modality (e.g., RDCT instead of “no imaging”) or presented both options (e.g., “POCUS or no imaging”). The Consensus guidelines universally recommended ultrasound for the youngest group and for pregnant patients, likely driven by Safety concerns to avoid ionizing radiation. For the oldest group, CT was the recommended initial imaging modality of choice, likely reflecting the concerns for in- creased risk of potentially serious Alternative diagnoses in this age group. For these three groups (youngest, pregnant, and oldest), the im- aging recommendation was straightforward. We separated out these special cases. Two groups remained: age 18-44 and age 45-64 years, for which we developed our decision tree.

1. Results

Our rules recommend ultrasound as the initial imaging modality for pediatric and pregnant patients and CT for older patients. We summa- rize the recommendations for these cases in Fig. 3. Our decision tree al- gorithm for non-pregnant patients aged 18-64 years first branched at age, separating the younger (18-44 years) from the older population (45-64 years), as in the consensus document. The second branch point in both subgroups was the presence of left lower quadrant

Fig. 4. Decision rules for ED patients aged 18-64 years with possible ureteral colic*. LLQ TTP = left lower quadrant of the abdomen tenderness to palpation; POCUS = point-of-care ultrasound; RDCT = reduced-dose computed tomography. ? The decision rules assume the following: recent onset of flank pain (<3h), nausea or vomiting, hematuria on urinalysis or point-of-care colorimetric urinalysis, and pain relief with intravenous analgesics.

tenderness. The final branch point of hydronephrosis on ultrasound in the younger group and history of kidney stones in the older group, led to the specific imaging recommendations. In the few cases where agree- ment in the consensus document was only moderate, we opted for the more conservative imaging approach or presented multiple options (e.g., ultrasound or no imaging). The final decision tree was simple and easy to follow and concordant with the consensus recommendations (Fig. 4).

1. Discussion

We developed clinical decision rules to guide imaging for ED pa- tients with suspected ureteral stones. Our starting point was the most recent available multispecialty consensus. We adhered to the consensus document wherever possible but made modifications as needed for clinical utility, feasibility, or simplicity. The resulting rules are simple, comprehensive, and easy to understand and apply to clinical practice.

Our clinical decision rules have the potential to safely limit CT use for ED patients with uncomplicated ureteral stones, particularly in younger patients. Although only a narrow subset of patients are likely to avoid CT use by using these rules, they are precisely the demographic that stands to gain the most benefit from reduced exposure to ionizing radiation at a younger age. These rules could be used to guide Imaging decisions for patients who present to the ED with acute pain suspected to be due to ureteral stone. In addition to the benefits to patient safety, these rules could potentially expedite ED throughput and save resources by reduc- ing low-yield CT imaging.

1. Limitations

There are several limitations to our study. The multispecialty con- sensus on which our rules are based was intended to provide recom- mendations on appropriate imaging, not to predict uncomplicated ureteral stone. In extrapolating clinical decision rules, we are going be- yond the scope of the original paper. This required modification of some

of the variables to reflect real-world ED conditions and to facilitate data collection if the rules are to be validated. Finally, the rules will require retrospective and prospective validation in varioUS settings before they can be routinely adopted in clinical practice (Fig. 2).

1. Conclusion

Using a recent multispecialty consensus as our starting point, we were able to create a simple set of clinical decision rules to guide diag- nostic imaging in ED patients with suspected ureteral colic. These rules have the potential to safely reduce the use of CT, particularly in younger patients.

Funding

Funding was provided by the Kaiser Permanente Northern California Community Benefit Program.

Credit author statement

The authors contributed as follows:

Both authors (EJD and DRV) contributed to the conceptualization, formal analysis, investigation, methodology, and writing and editing of the manuscript.

Conflict of interest statement

The authors, Edward J Durant and David R Vinson, have no conflicts of interest to disclose.

References

1. Hyams ES, Korley FK, Pham JC, Matlaga BR. Trends in imaging use during the emer- gency department evaluation of flank pain. J Urol. 2011;186(6):2270-4.
2. Fwu C-W, Eggers PW, Kimmel PL, Kusek JW, Kirkali Z. Emergency department visits, use of imaging, and drugs for urolithiasis have increased in the United States. Kidney Int. 2013;83(3):479-86.
3. Faulkner K, Moores B. radiation dose and somatic risk from computed tomography. Acta Radiol. 1987;28:483-8.
4. Brenner D, Elliston C, Hall E, Berdon W. Estimated risks of radiation-induced fatal cancer from pediatric CT. Am J Roentgenol. 2001;176(2):289-96.
5. Smith-Bindman R, Miglioretti DL, Larson EB. Rising use of diagnostic medical imag- ing in a large integrated health system. Health Aff (Millwood). 2008;27(6): 1491-502.
6. Wang RC, Rodriguez RM, Moghadassi M, et al. External validation of the STONE score, a clinical prediction rule for ureteral stone: an observational multi- institutional study. Ann Emerg Med. 2016;67(4):423-432.e422.
7. Christopher L Moore SBBDSLAMDSCPG. Derivation and validation of a clinical pre- diction rule for uncomplicated ureteral stone-the STONE score: retrospective and prospective Observational cohort studies. BMJ. 2014:1-12.
8. Wang RC, Rodriguez RM, Fahimi J, et al. Derivation of decision rules to predict clin- ically important outcomes in acute flank pain patients. Am J Emerg Med. 2017;35 (4):554-63.
9. Moore CL, Carpenter CR, Heilbrun ME, et al. Imaging in suspected renal colic: sys- tematic review of the literature and multispecialty consensus. Ann Emerg Med. 2019;74(3):391-9.
10. Laupacis A, Sekar N, Stiell I. Clinical prediction rules. A review and suggested modi-

fications of methodological standards. JAMA. 1997;277(6):488-94.