Omadacycline-Associated Pancreatic Injury: a Spectrum from Mild Interstitial to Recrudescent Pancreatitis with Pharmacovigilance Implications Authors: Guanghui Chen, Qiang Qu, Lishan Yi, Yanyan Zhou, Jian Qu Published: 2026 ABSTRACT IntroductionOmadacycline is a novel aminomethylcycline antibiotic with an incompletely characterized safety profile, particularly regarding pancreatic toxicity. A single case of omadacycline-induced acute pancreatitis has been reported, yet multiple FDA Adverse Event Reporting System (FAERS) analyses have failed to detect a pancreatitis signal. We aimed to characterize the spectrum of omadacycline-associated pancreatic injury and to explore the pharmacovigilance implications of cases presenting with subtle or atypical imaging features.MethodsWe report three cases from an intensive care unit in which patients developed pancreatic enzyme elevation (> 3 × ULN) temporally associated with omadacycline. Alternative etiologies were systematically excluded. Causality was assessed using the Naranjo scale. A literature search of PubMed and Embase was also performed.ResultsAll three patients exhibited predominant hyperlipasemia with CT-confirmed pancreatic injury spanning a severity spectrum. Case 1 (79F) showed lipase > 34 × ULN, peritoneal signs, and new pancreatic head enlargement with peripancreatic stranding (Naranjo 7); Case 2 (18M) had asymptomatic lipase elevation with diffuse glandular swelling and heterogeneous parenchymal density (Naranjo 5); Case 3 (73F) developed recrudescent focal pancreatitis during continued exposure, with tail involvement and perirenal fascial thickening (Naranjo 4). All improved after discontinuation. Common etiologies were excluded. The previously reported case showed an inverse pattern (imaging-dominant with mild enzyme elevation). Collectively, these four cases support upgrading omadacycline from Badalov Class IV to Class III.ConclusionsOmadacycline can cause a spectrum of pancreatic injury, from mild interstitial pancreatitis to recrudescent disease, with CT findings that may be unobtrusive and easily overlooked. This underrecognition likely contributes to underreporting, which may explain the absence of a FAERS pancreatitis signal. Clinicians should monitor lipase and amylase, maintain a low threshold for CT imaging when enzymes rise, and report suspected cases using specific MedDRA terms. Key Summary Points This first case series of omadacycline-associated pancreatic injury reveals a spectrum of unobtrusive or atypical imaging findings that may be easily missed on routine CT interpretationMultiple independent FDA Adverse Event Reporting System analyses found no pancreatitis signal, likely because mild or atypical imaging presentations evade standard diagnostic criteria and radiological recognition, leading to systematic underreportingMonitor amylase and lipase during therapy; unobtrusive CT abnormalities do not exclude clinically significant pancreatic injury. Report such cases with specific MedDRA terms to strengthen pharmacovigilance This first case series of omadacycline-associated pancreatic injury reveals a spectrum of unobtrusive or atypical imaging findings that may be easily missed on routine CT interpretationMultiple independent FDA Adverse Event Reporting System analyses found no pancreatitis signal, likely because mild or atypical imaging presentations evade standard diagnostic criteria and radiological recognition, leading to systematic underreportingMonitor amylase and lipase during therapy; unobtrusive CT abnormalities do not exclude clinically significant pancreatic injury. Report such cases with specific MedDRA terms to strengthen pharmacovigilance Introduction Severe pneumonia frequently necessitates intensive care unit (ICU) admission, in which infections caused by multidrug-resistant pathogens pose a significant therapeutic challenge [1]. Omadacycline, a novel aminomethylcycline approved in 2018 for community-acquired bacterial pneumonia and acute bacterial skin and skin structure infections [2], exhibits activity against resistant organisms including carbapenem-resistant Klebsiella pneumoniae and is increasingly used off-label when standard options fail [3]. Drug-induced pancreatitis (DIAP) in critically ill patients is frequently a diagnosis of exclusion, and delayed recognition may lead to recurrent injury and worse outcomes [4, 5]. Phase III trials reported gastrointestinal adverse events, but no cases of acute pancreatitis, with postbaseline lipase changes comparable to those of active comparators [6]. Nevertheless, the prescribing information carries a class-wide tetracycline warning for pancreatitis [7]. The Badalov classification system grades drugs by the strength of evidence for pancreatitis, from Class I (confirmed rechallenge) to Class IV (single case reports) [4]. Xu et al. reported the first case of acute pancreatitis possibly induced by omadacycline [8]. To date, multiple independent analyses of the FDA Adverse Event Reporting System (FAERS) (including one encompassing 4541 cases) have consistently failed to detect a signal of disproportionality for pancreatitis, instead identifying hepatic and gastrointestinal signals [9–11]. This discrepancy raises the possibility that omadacycline may cause a form of pancreatic injury that presents with atypical imaging features, thereby evading both radiological recognition and pharmacovigilance detection. We present three cases illustrating this phenomenon, with hyperlipasemia as the predominant feature and CT findings that were initially unrecognized until expert re-review confirmed a spectrum of interstitial pancreatic injury. Methods The three cases were identified retrospectively from patients who received omadacycline in our ICU between June and July 2026. All three patients developed pancreatic enzyme elevations (> 3 × ULN) temporally associated with omadacycline, and DIAP was suspected. Medical records were reviewed for clinical characterization and causality assessment. Alternative etiologies, including biliary disease, alcohol, hypertriglyceridemia, and trauma, were excluded by clinical history, laboratory data, and imaging. Naranjo scoring was performed by one investigator and independently verified by another author. All CT images were independently re-reviewed by an experienced abdominal radiologist, who was blinded to the clinical information and the initial radiology reports. Data Source Demographic data, clinical presentations, laboratory parameters (including serial amylase and lipase levels), imaging findings, concomitant medications, and clinical outcomes were retrieved from electronic medical records. Causality Assessment Causality was assessed using the Naranjo adverse drug reaction probability scale [12], focusing on the temporal relationship between omadacycline exposure and pancreatic enzyme abnormalities. Scores are interpreted as: definite (≥ 9), probable (5–8), possible (1–4), and doubtful (≤ 0). The findings were further interpreted within the Badalov classification framework for DIAP [4]. Bibliographic Research A systematic literature search of PubMed and Embase was conducted from inception through June 2026 using the combination of the terms "omadacycline" OR "NUZYRA" with "pancreatitis," "hyperamylasemia," "hyperlipasemia," "lipase," or "pancreatic injury." Reference lists of retrieved articles were also screened for additional cases. Ethics The study was conducted in accordance with the Declaration of Helsinki. The study protocol was approved by the Ethics Committee of the Second Xiangya Hospital of Central South University (LYEC2024-K0181). Written informed consent for publication of clinical details was obtained from all patients or their legal guardians. Results All three patients exhibited predominant hyperlipasemia with CT-confirmed pancreatic abnormalities on expert re-review, spanning a spectrum from mild focal interstitial pancreatitis to diffuse glandular involvement and recrudescent focal disease (Table 1). The temporal relationships among omadacycline exposure, enzyme elevation, imaging findings, and clinical recovery are summarized in Fig. 1. Alternative etiologies were excluded, and Naranjo scores ranged from 4 to 7. Table 1Clinical characteristics of three cases of omadacycline-associated pancreatic injuryFeatureCase 1 (79F)Case 2 (18M)Case 3 (73F)Primary diagnosesSepsis, pulmonary infection, diabetic foot, pressure ulcerPulmonary infection, fulminant T1DM, DKASevere pneumonia, chronic renal insufficiency, GI bleedingRace/ethnicityHan ChineseHan ChineseHan ChineseOmadacycline regimen200 mg IV load, 100 mg IV qd (day 4–8)200 mg IV load, 100 mg IV qd (day 1–4)200 mg IV load, 100 mg IV qd (day 1–10)Baseline lipase/amylasea19/23 U/L26.1/80 U/L382.6/167 U/LPeak lipase/amylase2062.3/714 U/L (day 8)1427.4/330 U/L(day 3)2089.4/601 U/L (day 10)Abdominal symptomsTenderness, rebound tendernessNoneSubxiphoid painLipase/amylase patternLipase-dominantLipase-dominantBiphasicKey CT findingbDay 1: normal.Day 8: pancreatic head enlargement, peripancreatic stranding and exudate (Fig. 2)Day 3: diffuse glandular swelling, heterogeneous density, peripancreatic exudate.Day 10: near-complete resolution (Fig. 3)Day 10: plump pancreatic tail with exudate, left anterior renal fascial thickening.Day 20: improvement (Fig. 4)OutcomeComplete normalizationMarked enzyme declineReturn to baselineNaranjo score7 (probable)5 (probable)4 (possible)DKA diabetic ketoacidosis, T1DM Type 1 diabetes mellitus, GI gastrointestinal, ULN upper limit of normalaInstitutional reference ranges: lipase 13–60 U/L, amylase 35–135 U/LbAll CT findings confirmed on independent re-review by an abdominal radiologistFig. 1Timeline of clinical course for three cases of omadacycline-associated pancreatic injury. The figure illustrates the temporal sequence of omadacycline administration, pancreatic enzyme elevation, imaging abnormalities, and clinical improvement in each case. In Case 3, symptom recrudescence occurred during continued omadacycline exposure, followed by resolution after discontinuation Table 1Clinical characteristics of three cases of omadacycline-associated pancreatic injuryFeatureCase 1 (79F)Case 2 (18M)Case 3 (73F)Primary diagnosesSepsis, pulmonary infection, diabetic foot, pressure ulcerPulmonary infection, fulminant T1DM, DKASevere pneumonia, chronic renal insufficiency, GI bleedingRace/ethnicityHan ChineseHan ChineseHan ChineseOmadacycline regimen200 mg IV load, 100 mg IV qd (day 4–8)200 mg IV load, 100 mg IV qd (day 1–4)200 mg IV load, 100 mg IV qd (day 1–10)Baseline lipase/amylasea19/23 U/L26.1/80 U/L382.6/167 U/LPeak lipase/amylase2062.3/714 U/L (day 8)1427.4/330 U/L(day 3)2089.4/601 U/L (day 10)Abdominal symptomsTenderness, rebound tendernessNoneSubxiphoid painLipase/amylase patternLipase-dominantLipase-dominantBiphasicKey CT findingbDay 1: normal.Day 8: pancreatic head enlargement, peripancreatic stranding and exudate (Fig. 2)Day 3: diffuse glandular swelling, heterogeneous density, peripancreatic exudate.Day 10: near-complete resolution (Fig. 3)Day 10: plump pancreatic tail with exudate, left anterior renal fascial thickening.Day 20: improvement (Fig. 4)OutcomeComplete normalizationMarked enzyme declineReturn to baselineNaranjo score7 (probable)5 (probable)4 (possible)DKA diabetic ketoacidosis, T1DM Type 1 diabetes mellitus, GI gastrointestinal, ULN upper limit of normalaInstitutional reference ranges: lipase 13–60 U/L, amylase 35–135 U/LbAll CT findings confirmed on independent re-review by an abdominal radiologist Clinical characteristics of three cases of omadacycline-associated pancreatic injury Day 1: normal. Day 8: pancreatic head enlargement, peripancreatic stranding and exudate (Fig. 2) Day 3: diffuse glandular swelling, heterogeneous density, peripancreatic exudate. Day 10: near-complete resolution (Fig. 3) Day 10: plump pancreatic tail with exudate, left anterior renal fascial thickening. Day 20: improvement (Fig. 4) DKA diabetic ketoacidosis, T1DM Type 1 diabetes mellitus, GI gastrointestinal, ULN upper limit of normal aInstitutional reference ranges: lipase 13–60 U/L, amylase 35–135 U/L bAll CT findings confirmed on independent re-review by an abdominal radiologist Timeline of clinical course for three cases of omadacycline-associated pancreatic injury. The figure illustrates the temporal sequence of omadacycline administration, pancreatic enzyme elevation, imaging abnormalities, and clinical improvement in each case. In Case 3, symptom recrudescence occurred during continued omadacycline exposure, followed by resolution after discontinuation Case 1: Acute Interstitial Pancreatitis with Peritoneal Signs A 79-year-old female with sepsis, pulmonary infection, diabetic foot, and an infected sacral pressure ulcer was admitted on day 1 with a normal abdominal CT (Fig. 2a). Omadacycline was initiated on day 4 (200 mg IV load, 100 mg IV qd), when baseline lipase and amylase were 19 U/L and 23 U/L, respectively.Fig. 2Abdominal CT of Case 1. a Axial image on day 1 (admission, prior to omadacycline therapy) shows a normal pancreas without peripancreatic stranding or exudate. b Axial image on day 8 (after 5 days of omadacycline therapy) demonstrates newly developed pancreatic head enlargement with surrounding peripancreatic fat stranding and exudative changes, consistent with acute interstitial pancreatitis Abdominal CT of Case 1. a Axial image on day 1 (admission, prior to omadacycline therapy) shows a normal pancreas without peripancreatic stranding or exudate. b Axial image on day 8 (after 5 days of omadacycline therapy) demonstrates newly developed pancreatic head enlargement with surrounding peripancreatic fat stranding and exudative changes, consistent with acute interstitial pancreatitis On day 8, lipase surged to 2062.3 U/L (> 34 × ULN) and amylase to 714 U/L, accompanied by abdominal and rebound tenderness. CT on day 8 revealed newly developed pancreatic head enlargement with peripancreatic fat stranding and exudative changes (Fig. 2b), consistent with acute interstitial pancreatitis. Although cholelithiasis was noted on biliary imaging, the normal baseline enzymes and the absence of biliary duct dilatation excluded a biliary etiology. Omadacycline was discontinued on day 8. By day 10, lipase had fallen to 64.9 U/L, amylase to 26 U/L, and peritoneal signs resolved. The Naranjo score was 7 (probable). Case 2: Diffuse Interstitial Pancreatitis with Metabolic Confounders An 18-year-old male with fulminant type 1 diabetes (FT1DM) and diabetic ketoacidosis (DKA) received omadacycline (200 mg IV load, 100 mg IV qd). Despite remaining asymptomatic, lipase surged from a baseline of 26.1 U/L to 1427.4 U/L, and amylase increased from 80 U/L to 330 U/L by day 3 of therapy. Same-day CT showed mild swelling of the pancreatic head, body, and tail with heterogeneous parenchymal density and peripancreatic exudates (Fig. 3a).Fig. 3Abdominal CT of Case 2. a Axial image on day 3 of omadacycline therapy shows mild diffuse swelling of the pancreatic head, body, and tail with heterogeneous parenchymal density and peripancreatic exudates. b Follow-up CT on day 10 demonstrates near-complete resolution of peripancreatic exudates Abdominal CT of Case 2. a Axial image on day 3 of omadacycline therapy shows mild diffuse swelling of the pancreatic head, body, and tail with heterogeneous parenchymal density and peripancreatic exudates. b Follow-up CT on day 10 demonstrates near-complete resolution of peripancreatic exudates Omadacycline was discontinued on day 4. Follow-up CT on day 10 (Fig. 3b) showed near-complete resolution of peripancreatic exudates, with enzymes declining steadily to lipase 72.8 U/L and amylase 100 U/L by day 17. Renal function recovered progressively (eGFR improved from 48 mL/min on admission to 86 mL/min by day 10), excluding renal dysfunction as a cause of pseudohyperamylasemia. Despite FT1DM and DKA as confounders, the marked lipase predominance, temporal relationship with drug withdrawal, and CT evidence of diffuse pancreatitis all supported a contributory role for omadacycline. The Naranjo score was 5 (probable). Case 3: Recrudescent Focal Pancreatitis on Continued Therapy A 73-year-old female with severe pneumonia, chronic renal insufficiency, and gastrointestinal bleeding presented with subclinical pancreatic injury (admission amylase 167 U/L, lipase 382.6 U/L; no abdominal symptoms). Omadacycline was started (200 mg IV load, 100 mg IV qd). On day 6, she developed subxiphoid pain with lipase rising to 1048.8 U/L and amylase to 315 U/L. Somatostatin was administered and improved symptoms, but omadacycline was continued. On day 10, pain recurred with lipase surging to 2089.4 U/L and amylase to 601 U/L. CT revealed a plump pancreatic tail with peripancreatic exudate and left anterior renal fascial thickening (Fig. 4a, c). Omadacycline was discontinued.Fig. 4Abdominal CT of Case 3. a Axial image on day 10 of omadacycline therapy demonstrates a plump pancreatic tail with surrounding exudate. b Axial image on day 20 shows decreased peripancreatic exudate. c Axial image on day 10 at the level of the left kidney demonstrates left anterior renal fascial thickening. d Follow-up CT on day 20 shows reduced left anterior renal fascial thickening Abdominal CT of Case 3. a Axial image on day 10 of omadacycline therapy demonstrates a plump pancreatic tail with surrounding exudate. b Axial image on day 20 shows decreased peripancreatic exudate. c Axial image on day 10 at the level of the left kidney demonstrates left anterior renal fascial thickening. d Follow-up CT on day 20 shows reduced left anterior renal fascial thickening Follow-up CT on day 20 (Fig. 4b, d) showed improvement, and enzymes had declined to lipase 341.1 U/L and amylase 162 U/L with complete pain resolution. The sequence of recrudescence during continued exposure and resolution after withdrawal, though confounded by baseline pancreatic injury, was consistent with omadacycline-associated injury. The Naranjo score was 4 (possible). Literature Review The literature search identified only one prior case report of omadacycline-associated pancreatitis [8], which presented with CT-confirmed exudation but only modest enzyme elevation. Three FAERS-based pharmacovigilance studies reported no signal of pancreatitis for omadacycline [9–11]. According to the Badalov classification [4], a single case report designates a drug as Class IV; two or more cases without rechallenge or consistent latency elevate a drug to Class III. The prior case placed omadacycline in Class IV. Our three cases thus meet the criteria for Class III designation. Discussion We present three cases of omadacycline-associated pancreatic injury unified by predominant hyperlipasemia and objective CT abnormalities, findings that were unobtrusive on preliminary interpretation and confirmed on expert re-review. To our knowledge, this is the first case series to describe this phenotypic diversity and its implications for the observed absence of the FAERS pancreatitis signal. DIAP accounts for 2–5% of acute pancreatitis cases yet remains notoriously underdiagnosed [5], owing to the absence of mandatory reporting, clinician bias, difficulty excluding common etiologies in comorbid patients, and a frequent lack of rechallenge or defined latency to confirm causality [4]. Tetracycline-class antibiotics are well-recognized causes of DIAP, with cases reported for tigecycline, minocycline, doxycycline, and eravacycline [8]. The mechanisms remain incompletely understood but likely involve direct cytotoxicity, hypersensitivity reactions, or toxic effects of an unknown metabolite [13]. A key finding is that all three patients had objective CT evidence of pancreatic injury, yet the imaging abnormalities escaped initial detection. These patterns represent mild interstitial pancreatitis, which is inherently less conspicuous than the necrosis or fluid collections of severe disease. Contrast-enhanced CT misses 10–20% of cases of mild pancreatitis [14, 15], and even advanced techniques such as dual-energy CT with radiomics-based analysis still face diagnostic challenges in this setting [16]. In critically ill patients with multiple competing CT abnormalities, such subtle findings are particularly susceptible to underrecognition. This diagnostic difficulty has direct consequences for pharmacovigilance. In FAERS, adverse events are coded using MedDRA Preferred Terms [17]; when CT findings are reported with non-committal language, clinicians may not code the event as "pancreatitis," instead using non-specific terms such as "hyperlipasemia" that are invisible to pancreatitis-specific disproportionality analyses. These terms are structurally excluded from disproportionality analyses that scan specifically for the pancreatitis Preferred Term, effectively diluting or masking any true signal. This systematic misclassification offers a plausible explanation for the consistent absence of a pancreatitis signal for omadacycline [9–11]. A consistent finding across all three patients was predominant hyperlipasemia. Lipase is more specific than amylase for pancreatic acinar injury [18], and its predominance strengthened the case for drug-induced injury, including in Case 2, where the lipase elevation far exceeded the modest changes typical of FT1DM or DKA [19, 20]. Rapid enzyme decline following withdrawal further reinforced the causal association. Therefore, lipase should be the primary biomarker for monitoring during omadacycline therapy. Case 3 adds a critical safety dimension, demonstrating that continued omadacycline exposure during suspected pancreatic injury can provoke recrudescent pancreatitis. Somatostatin masked ongoing injury; when omadacycline was continued, pain and elevated enzymes recurred. This sequence reinforces a fundamental principle: prompt withdrawal of the offending agent remains the cornerstone of managing suspected drug-induced pancreatitis [5, 13]. The application of the Badalov classification to this case series warrants careful consideration. These three cases exhibit heterogeneous latency and lack rechallenge. Nevertheless, they collectively strengthen the cumulative evidence: Case 1 demonstrates a clear temporal sequence with normal baseline enzyme levels, Case 2 provides objective CT evidence despite metabolic confounders, and Case 3 illustrates a recrudescence pattern with continued exposure. This triangulation, encompassing different baseline states and clinical presentations, provides provisional support for a Class III designation under a conservative interpretation of the Badalov framework, although larger prospective studies are needed to confirm this classification. This case series adds to the literature by documenting three patterns of omadacycline-associated pancreatic injury: enzyme-imaging dissociation, injury despite metabolic confounders, and recrudescence on continued exposure. These findings offer a mechanistic explanation for the absence of a FAERS signal and underscore the importance of clinical vigilance, prompt enzyme monitoring, and early drug discontinuation. The potential risk of pancreatic injury associated with omadacycline must be weighed against its clinical benefit on a case-by-case basis. This applies both to its approved indications, such as community-acquired pneumonia, and to off-label use in multidrug-resistant infections. In patients receiving omadacycline, particularly those with pre-existing pancreatic disease or risk factors, clinicians should maintain heightened vigilance and a low threshold for discontinuing the drug if enzyme elevation or abdominal symptoms develop [5, 13]. Limitations This study has several limitations. First, it is a retrospective case series with a small sample size from a single center, which limits generalizability. Second, although common alternative etiologies were systematically excluded, the multifactorial organ dysfunction and polypharmacy inherent to critical illness mean that residual confounding cannot be entirely ruled out. Third, rechallenge was not performed for ethical reasons. Fourth, as with all case series, these cases may reflect the most severe or most readily recognized presentations, and the absence of a denominator precludes estimation of incidence; larger prospective studies are needed to confirm our observations. Finally, contrast-enhanced CT protocols were not uniformly applied, which may have reduced sensitivity for subtle parenchymal changes. Conclusions Atypical CT findings in omadacycline-associated pancreatic injury are easily overlooked, driving underrecognition that may explain the absent FAERS signal. These three cases, together with the previously reported case, provide provisional support for upgrading omadacycline from Badalov Class IV to Class III, though confirmation from larger prospective studies is needed. Clinicians should monitor amylase and lipase, maintain a low threshold for CT imaging, and recognize that even unobtrusive abnormalities may represent clinically significant injury.