a b s t r a c t

procalcitonin is a biomarker with greater specificity for bacterial infection than other current laboratory markers. However, PCT can also be elevated in the setting of several nonInfectious conditions. A recent case re- port describes a patient with elevated PCT in the context of acute methamphetamine intoxication, but without evidence of infection. Thus far, no studies have evaluated the diagnostic utility of PCT in patients with active methamphetamine use. We seek to test the hypothesis that PCT has diminished utility in patients who use meth- amphetamine presenting to the Emergency Department (ED). We performed a retrospective cohort study of pa- tients presenting to an academic ED between May 2017 and July 2019. We included patients >=18 years of age with a positive urine methamphetamine test and at least two PCT results. Pregnant patients were excluded. Cases were classified as microbiologically documented infection, clinically documented infection, possible infec- tion, or no infection by clinician review. A positive PCT value was defined as >=0.5 ng/ml. The performance of PCT as a diagnostic test for bacterial infection in this population was then evaluated using sensitivity, specificity, false positive rate, false negative rate, and area under the receiver operating characteristic curve. We identified 143 pa- tients, including 75 with recorded PCT levels >=0.5 ng/ml and 93 with microbiologically or clinically documented bacterial infection. The sensitivity and specificity of PCT for bacterial infection in this study population was 60% and 64%, respectively. The false positive rate was 36% while the false negative rate was 40%. The area under the ROC curve was 0.65. Additionally, we describe 8 patients with confirmed absence of infection but with ele- vated PCT, 4 of whom had serum values >10 ng/ml. The results suggest that PCT has poor diagnostic utility for bacterial infection in patients with active methamphetamine use presenting to the ED.

(C) 2022

1. Introduction

The Biomarker procalcitonin is a potential tool in the diagnosis and management of patients with bacterial infections. Under physio- logic conditions PCT is secreted by thyroid C-cells, where it is cleaved into calcitonin, which is involved in calcium homeostasis [1]. In bacterial infection, however, various non-endocrine tissues also secrete PCT [1,2]. Serum PCT levels have been shown to rise 2-4 h following inflammatory stimuli and have a half-life of 24 to 36 h [1]. These kinetics coupled with PCT’s bacterial specificity make it a potential tool in the initiation of an- tibiotics in patients with suspected infection, as well as the de- escalation of antibiotics in patients showing clinical improvement [3-5].

* Corresponding author at: 200 West Arbor Drive, San Diego, CA 92103, United States of America.

E-mail addresses: bkennis@health.ucsd.edu (B. Kennis), a2ali@health.ucsd.edu (A. Ali), dlasoff@health.ucsd.edu (D. Lasoff), dasweeney@health.ucsd.edu (D.A. Sweeney), gwardi@health.ucsd.edu (G. Wardi).

However, several noninfectious conditions that limit the diagnostic utility of PCT have been identified, including severe trauma, burns, sur- gery, and pancreatitis [5-7]. Patients with kidney and Liver failure have also been shown to have elevated serum PCT levels in the absence of in- fection due to impaired clearance [8,9]. Knowledge of these non- infectious causes of elevated PCT levels aids in the more accurate use and interpretation of test results in the Emergency Department (ED).

Methamphetamine toxicity may represent a previously unidentified confounder of PCT testing. A recent case report described a patient with acute methamphetamine intoxication without evidence of bacterial in- fection, yet the patient’s serum PCT level was profoundly elevated (1640 ng/ml, normal <0.07 ng/ml) [10]. Methamphetamine has been shown to have various effects on human immunity and inflammation, but its effect on PCT levels has not been systematically described [11]. More- over, the performance characteristics of PCT testing in the setting of methamphetamine toxicity is not known.

https://doi.org/10.1016/j.ajem.2022.01.049

0735-6757/(C) 2022

Individuals presenting to the ED with methamphetamine use is in- creasingly common. In 2016, 3.2% of all ED visits in the United States were associated with a methamphetamine positive urine toxicology result [12]. At the same time, PCT testing is now frequently ordered in the ED setting to determine whether a patient is infected. Thus, it is important to understand how PCT testing for the presence of infection performs in patients with methamphetamine use. We hypothesize that procalcitonin has diminished diagnostic utility for bacterial infec- tion in the setting of methamphetamine use.

1. Methods

We performed a retrospective cohort study of patients admitted to an academic medical system between May 2017 and July 2019. We in- cluded patients >=18 years old presenting to the ED with a positive urine methamphetamine test and at least two PCT results. Our institu- tion’s urine toxicology protocol consists of a screening immunoassay followed by a confirmatory liquid chromatography with tandem mass spectrometry (LC/MS/MS). We included only patients with a positive confirmatory LC/MS/MS in this study. Pregnant patients were excluded. Data was abstracted via automatic query of the electronic medical re- cord and was approved by The University of California San Diego Human Research Protection Program. Cases were classified as microbi- ologically documented infection, clinically documented infection, possi- ble infection, or no infection by one of the study authors using criteria modified from previously validated methodology (Table 1) [13]. Follow- ing classification, 10% of cases were independently classified by a sec- ond study author and a Cohen’s kappa coefficient was calculated. Any discrepancies were then adjudicated by a third study author. Clinically or microbiologically confirmed infections were defined as true infec- tions and a positive PCT value was defined as >=0.5 ng/ml, the institu- tional value above which antibiotics are strongly recommended. The performance of PCT as a diagnostic test for bacterial infection in this population was evaluated using sensitivity, specificity, false positive rate, false negative rate, and area under the receiver operating charac- teristics (ROC) curve. Analysis of variance (ANOVA) was performed

comparing the mean maximum PCT across each infectious category. Statistical analysis was performed using SPSS 28.0 (IBM). Finally, cases with elevated PCT but absent infection or other causes of elevated PCT were qualitatively described.

1. Results

We identified 143 patients who tested positive for methamphet- amine and had >=2 documented PCT test results. The majority of these patients were men (78%) with a median age of 51.0 years. Approxi- mately half (49.7%) of the study patient population was admitted to the intensive care unit and had a mean peak Sequential Organ Failure Assessment score of 5.5. Seventy-five (52%) patients had at least one PCT value greater than or equal to 0.5 ng/ml while 93 (65%) were found to have confirmed infections. Fifty-six had microbiologically confirmed infection, 37 had clinically confirmed infection, 32 had no in- fection, and 18 had possible infection (Table 2). Cohen’s kappa coeffi- cient for the assignment of infectious category was 0.93, with a single case of the 15 reviewed by a second physician requiring adjudication by a third physician. Procalcitonin as a diagnostic indicator of microbio- logically or clinically confirmed bacterial infection in patients with a positive methamphetamine urine toxicology test had a sensitivity and specificity of 60% and 64%, respectively. The false positive rate was 36% while the false negative rate was 40%. The area under the ROC curve was 0.65 (Fig. 2). ANOVA showed no significant difference in mean maximum PCT values across the four infectious categories (p = 0.51) (Fig. 1). Furthermore, we identified 8 patients with confirmed ab- sence of bacterial infection with elevated PCT, four of whom had PCT values >10 ng/ml (Table 3).

1. Discussion

This represents the largest study to date assessing the diagnostic ac- curacy of serum PCT testing for the presence of infection in patients pre- senting to an emergency department with a concomitant positive methamphetamine urine toxicology. We found that the sensitivity and

Table 1

Infectious categories.

Microbiologically confirmed infection Clinically confirmed infection Absence of infection Possible

infection

Definition Presence of a clinical or radiological infectious focus, or both, and pathogen identification.

Categories were adapted from those used in the PRORATA trial [13].

Presence of a clinical or radiological infectious focus, or both, without causative pathogen identification.

Absence of a clinical or radiological infectious focus, or both.

All other scenarios.

Fig. 1. Maximum PCT by infectious category.

specificity of PCT for bacterial infection in this population was 60% and 64%, compared to 88% and 81% in prior studies of hospitalized patients suspected of having a bacterial infection [3]. While the published data to support the use of PCT to identify bacterial infection are conflicting [14], these results suggest that the accuracy of PCT testing to identify pa- tients with an infection is limited in the setting of methamphetamine toxicity. Thus, PCT values in patients with concomitant methamphet- amine toxicity should be interpreted with caution. Of note, this study was not designed to assess the utility of serial PCT testing for the pur- pose of de-escalating antibiotics.

Importantly, we identified 8 patients with positive methamphet- amine tests and elevated PCT levels but absent infection on clinical re- view (Table 3). It is important to note that these patients presented not only as false positives (absence of microbiologic or clinical

Fig. 2. ROC curve.

infection), but also specifically as not infected on clinician review (con- firmed absence of infection). There have been rare reports on unin- fected patients with extremely high PCT test results in the setting of methamphetamine toxicity. Previously, one case study described a pa- tient admitted to the ICU with methamphetamine toxicity, rhabdomy- olysis, multiorgan failure, and a serum PCT value of 1640 ng/ml but without evidence of infection [10]. In this study we identified 4 individ- uals with strikingly elevated PCT measures >10 ng/ml. We did not iden- tify an infectious or previously identified noninfectious cause for these extremely elevated PCT results. The patients in this case series pre- sented with no clear pattern of pathologies and comorbidities that could alternatively explain an elevated serum PCT level.

Of particular interest are two patients who presented with maxi- mum PCT values of 53.75 ng/ml and 96.64 ng/ml, respectively, following reported intravenous (IV) methamphetamine use. Both patients pre- sented to the ED with a chief complaint of head and neck pain, and were found to have tachycardia, elevated serum lactate, and leukocyto- sis, but no evidence of infection was identified. Neither of these patients had a previously identified noninfectious reason to have an elevated PCT test. One of these patients had multiple presentations to the ED with similar symptoms and laboratory markers but without discernible pathology. The fact that remarkably elevated PCT values were produced following IV methamphetamine use suggests a possible relationship be- tween either route of administration or dose of methamphetamine and serum PCT level.

While immunomodulatory and proinflammatory effects of metham- phetamine have been previously reported, a direct effect of metham- phetamine on serum PCT has yet to be suggested [11]. It is possible that methamphetamine stimulates inflammatory pathways leading to increased PCT measures. Tumor necrosis factor alpha (TNF-?), interleu- kin 6 (IL-6), and several other cytokines have been shown to increase expression of the CALC-1 gene that codes for PCT [15]. Previous studies have demonstrated higher circulating levels of TNF-? in patients who habitually use methamphetamine [16,17]. Animal models have sug- gested that methamphetamine mediated damage to the intestinal mu- cosa leads to increased circulating exotoxin and TNF-? [18]. Likewise, postmortem studies have also demonstrated damage to the blood brain barrier in patients with chronic methamphetamine use, which can further increase circulating TNF-? [19]. Another possibility is that methamphetamine potentiates the rise in serum PCT caused by other pathologies. Methamphetamine has been previously shown to be both hepatotoxic and nephrotoxic [20]. Similarly, elevated PCT has been ob- served in both patients with liver and kidney failure due to decreased

Table 3

Case series of patients with elevated PCT in the absence of bacterial infection.

DM1: type 1 diabetes, DM2: type 2 diabetes, IDDM: insulin dependent diabetes, HCV: Hepatitis C virus, HIV: human immunodeficiency virus, HTN: hypertension, NICM: nonischemic car- diomyopathy, HFrEF: heart failure with reduced ejection fraction, COPD: chronic obstructive pulmonary disease, CKD: chronic kidney disease, MDD: major Depressive disorder, AUD: Alcohol use disorder

clearance [8,9]. It is possible that methamphetamine mediated damage of these organs might acutely decrease the clearance of PCT in the set- ting of other pathology that can induce its production. This could poten- tially explain the uncharacteristically high PCT values observed in our subjects. Finally, the relationship between methamphetamine and serum PCT could be confounded by other pathology associated with drug use, such as trauma or occult infection. Further research is needed to investigate these potential mechanisms.

There are also plausible mechanisms by which methamphetamine might suppress serum PCT levels. Methamphetamine has been previ- ously observed to impair elements of both adaptive and innate immunity [11]. It has likewise been associated with a more Rapid progression from human immunodeficiency virus to Acquired immunodeficiency syndrome (AIDS) [11]. It is conceivable that this immunosuppressive effect could also impair the induction of PCT production in response to bacterial infection. Furthermore, methamphetamine use is also associ- ated with HIV infection, which has been independently associated with lower serum PCT levels in the setting of bacterial infection compared to patients without HIV [21]. Methamphetamine thus has various immune-dysregulating associations to explain both the false positive and false negative PCT results observed in this study. The determination of which patients will exhibit which pattern of dysregulation with re- spect to PCT warrants further investigation.

There are limitations to this study. The first is that no standardized protocol for the collection of PCT was observed. While our institution provides guidelines and exclusion criteria regarding which patients are appropriate for PCT testing, it is likely that these protocols were not rigorously adhered to. Although this manner of collection may not be ideal, it is representative of how PCT testing is practically utilized across many healthcare systems [22]. The second is the inability to cor- relate the dose, time, or method of methamphetamine administration with serum PCT values. Consequently, it is unclear if a dose response re- lationship exists between methamphetamine and serum PCT. Third, it is important to emphasize the retrospective nature of this study. Con- founding is possible at the level of the patient and treatment team with this design, and it is vulnerable to incomplete and inaccurate doc- umentation. Similarly, PCT values were available during the process of assigning infectious categories, which was performed by the study au- thors. While this welcomes misclassification bias, its effect would be to affirm the diagnostic accuracy of PCT rather than to refute it, as this study does. Nonetheless, PCT was not in the criteria for assigning infec- tious categories and the inter-rater reliability for the fifteen cases re- viewed by two study authors demonstrated near perfect agreement (Cohen’s kappa coefficient 0.93). Finally, our analysis is a single center study with modest sample size and thus may lack generalizability. De- spite these limitations, this study suggests that the use of serum PCT

testing to diagnose bacterial infection has poor sensitivity and specific- ity in patients presenting to the ED with a concomitant positive urine methamphetamine toxicology. Further research is warranted to more thoroughly isolate and quantify the relationship between methamphet- amine and serum procalcitonin levels.

Funding

This project was partially supported by the National Institutes of Health, Grant UL1TR001442 of Clinical and Translational Science Award Funding.

Credit authorship contribution statement

Brent Kennis: Writing – review & editing, Writing – original draft, Software, Investigation, Formal analysis, Data curation. Daniel Lasoff: Writing – review & editing. Daniel A. Sweeney: Writing – review & editing, Validation, Supervision, Conceptualization. Gabriel Wardi: Writing – review & editing, Supervision, Project administration.

Declaration of Competing Interest

Dr. Wardi has been supported by funding from the National Founda- tion of Emergency Medicine, the Gordon and Betty Moore Foundation (GBMF9052), and the National Institutes of Health (R35GM143121), al- though unrelated to this manuscript. Mr. Kennis, Dr. Ali, Dr. Lasoff, and Dr. Sweeney have no disclosures to report. This project was partially supported by the National Institutes of Health, Grant UL1TR001442 of Clinical and Translational Science Award Funding.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2022.01.049.

References

1. Schneider Hans-Gerhard, Lam Que Thanh. Procalcitonin for the clinical laboratory: a review. Pathology. 2007;39(4):383-90.
2. Muller Beat, et al. Ubiquitous expression of the calcitonin-i gene in multiple tissues in response to sepsis. J Clin Endocrinol Metabol. 2001;86(1):396-404.
3. Simon Liliana, et al. Serum procalcitonin and C-reactive protein levels as markers of bacterial infection: a systematic review and meta-analysis. Clin Infect Dis. 2004;39 (2):206-17.
4. Schuetz Philipp, et al. Procalcitonin to guide initiation and duration of antibiotic treatment in acute Respiratory infections: an Individual patient data meta-analysis. Clin Infect Dis. 2012;55(5):651-62.
5. Davies Julie. Procalcitonin. J Clin Pathol. 2015;68(9):675-9.
6. Tang Benjamin MP, et al. Accuracy of procalcitonin for sepsis diagnosis in critically ill patients: systematic review and meta-analysis. Lancet Infect Dis. 2007;7(3):210-7.
7. Mofidi Reza, et al. The value of procalcitonin at predicting the severity of acute pan- creatitis and development of infected pancreatic necrosis: systematic review. Sur- gery. 2009;146(1):72-81.
8. Grace Eddie, Mackenzie Turner R. Use of procalcitonin in patients with various de- grees of chronic kidney disease including renal replacement therapy. Clin Infect Dis. 2014;59(12):1761-7.
9. Dong Ruolin, et al. Procalcitonin and liver disease: a literature review. J Clin Transl Hepatol. 2019;7(1):51.
10. Lovas Andras, et al. Extreme procalcitonin elevation without proven bacterial infec- tion related to amphetamine abuse. Case reports in critical care 2014; 2014.
11. Salamanca Sergio A, et al. Impact of methamphetamine on infection and immunity. Front Neurosci. 2015;8:445.
12. Richards John R, et al. Methamphetamine use and emergency department utiliza- tion: 20 years later. J Addict. 2017.;2017.
13. Bouadma Lila, et al. Use of procalcitonin to reduce patients’ exposure to antibiotics in intensive care units (PRORATA trial): a multicentre randomised controlled trial. Lan- cet. 2010;375(9713):463-74.
14. Becker Kenneth L, Snider Richard, Nylen Eric S. Procalcitonin assay in systemic in- flammation, infection, and sepsis: clinical utility and limitations. Crit Care Med. 2008;36(3):941-52.
15. Vijayan Ashitha L, et al. Procalcitonin: a promising diagnostic marker for sepsis and antibiotic therapy. J Intensive Care. 2017;5(1):1-7.
16. Wei Ze-Xu, et al. Aberrations in peripheral inflammatory cytokine levels in sub- stance use disorders: a meta-analysis of 74 studies. Addiction. 2020;115(12): 2257-67.
17. Luo Yayan, et al. Elevated serum levels of TNF-?, IL-6, and IL-18 in chronic metham- phetamine users. Hum Psychopharmacol Clin Experiment. 2021.;37:e2810.
18. Shen Simin, et al. Methamphetamine-induced alterations in Intestinal mucosal bar- rier function occur via the microRNA-181c/TNF-?/tight junction axis. Toxicol Lett. 2020;321:73-82.
19. Moretti Matteo, et al. Drug abuse-related neuroinflammation in human postmortem brains: an immunohistochemical approach. J Neuropathol Exp Neurol. 2019;78(11): 1059-65.
20. Carvalho Marcia, et al. Toxicity of amphetamines: an update. Arch Toxicol. 2012;86 (8):1167-231.
21. Mikula Tomasz, Cianciara Janusz, Wiercinska-Drapalo Alicja. Is there any influence of immune deficit on procalcitonin results? Hum Immunol. 2011;72(12):1194-7.
22. Hohn Andreas, et al. Adherence to a procalcitonin-guided antibiotic treatment pro- tocol in patients with severe sepsis and septic shock. Ann Intensive Care. 2018;8 (1):1-10.