Diffuse pericardial, intraabdominal, and subcutaneous non-infected fat necrosis in patient with chronic pancreatitis, managed with thoracoscopic and abdominal debridement

Abstract

We describe a patient that acutely presented with concomitant non-infected diffuse pericardial, intraabdominal, and subcutaneous fat necrosis in the setting of previously undiagnosed chronic calcific pancreatitis. There was no associated pancreatic necrosis, pseudocyst, or fistula. We have not encountered in the literature a previous description. The patient was managed with combination of abdominal and thoracoscopic resections and a favorable 4-year outcome. Exact mechanism of this case is uncertain. Based on clinical, intraoperative, and imaging findings, we consider that the etiology was intraabdominal and most likely related to chronic pancreatitis in the setting of metabolic syndrome.

Introduction

Retroperitoneal and intraabdominal fat necrosis in the omentum, mesentery, and falciform ligament is a known complication of pancreatitis. Extra-pancreatic necrosis without pancreatic parenchymal necrosis (EXPN) is an uncommon emerging entity accounting for 6–10% of interstitial pancreatitis cases^[1,2^]. Gallstones and alcohol are responsible for most of EXPN cases, while in approximately 20 % of patients the etiology is unknown.

Extra-abdominal (e.g., subcutaneous panniculitis and intraosseous) fat necrosis is extremely rare^[3^]. Other etiologies for systemic or diffuse fat necrosis can include vasculitis associated with lupus and erythema nodosum, or poststreptococcal vasculitis events. Alpha-1 antitrypsin deficiency is the most common inborn metabolic issue causing panniculitis.

HIGHLIGHTS

• Diffuse pericardial fat necrosis associated with pancreatitis and managed with thoracoscopic debridement—a case report.

• Pericardial fat necrosis associated with pancreatitis.

• Thoracoscopic debridement of pericardial fat necrosis.

• Mediastinal necrosis.

• Pancreatic panniculitis.

Outside of adult respiratory distress syndrome and pleural effusions, intrathoracic complications of acute pancreatitis are not common^[4,5^]. Rare reports of septic ascending mediastinitis are essentially all related to infected pancreatic necrosis, pancreatic duct disruption, fistulae, and pseudocysts extending into the chest. These cases are associated with high morbidity and mortality and require aggressive debridement and duct drainage interventions for patients to survive^[6–10^]. Thoracoscopic drainage of bilateral empyemas and mediastinitis associated with the pancreatic pseudocyst has been reported^[11^].

The exact mechanism of fat necrosis distant to pancreas in the absence of pancreatic fistula is not clear. We have not encountered descriptions of concomitant diffuse non-infected pericardial, intraabdominal, and subcutaneous fat necrosis. We suspect the etiology of this condition is chronic pancreatitis, possibly with mild acute exacerbation, in the setting of metabolic syndrome.

The Institutional Review Board of the University of Iowa considers case reports to be exempt from the application process. The patient provided consent for publication.

Clinical summary

A 49-year-old female presented with a 1-day history of acute onset of abdominal pain with no prior documented pancreatitis events. The patient had a BMI of 46.9 (weight 120 kg, height 160 cm), was a smoker, and a regular weekend alcohol user.

On initial presentation to local hospital, the patient was afebrile with blood pressure 122/70 mm Hg and heart rate 120 bpm. White blood cell count (WBC) was 13.7 K/mm³, alkaline phosphatase 137 U/L, alanine aminotransferase 239 U/L, aspartate aminotransferase 184 U/L, calcium 7.7 mg/dl, fasting triglycerides 295 mg/dL, and high-density lipids 26 mg/dL. Bilirubin, lactic acid, creatinine, and other parameters were within normal limits. Initial elevated lipase of 53 U/L increased to 64 U/L 8 hours later and normalized to 43 U/L in 16 hours (reference range 22–51 U/L). Computed tomography (CT) of the abdomen demonstrated edema of the falciform ligament and signs of chronic atrophic pancreatitis without evidence of peripancreatic necrosis and small right pleural effusion (Fig. 1A). Given the presence of pericholecystic fluid, thickening of the gallbladder wall to 3 mm, albeit no gallstones, on ultrasound imaging, the patient was taken for laparoscopic cholecystectomy for presumed acute cholecystitis. During laparoscopy, a non-inflamed gallbladder was visualized with the finding of necrotic falciform ligament. An operation was aborted, and the intubated patient was transferred to our institution for further care.

Figure 1.

(A) Abdominal CT demonstrating inflammatory changes of the preperitoneal fat and falciform ligament. Viable atrophic pancreas has a dilated duct and calcifications. Note absence of peripancreatic inflammation and fluid collections. (B) Necrosis of the falciform ligament and preperitoneal fat was encountered during laparotomy. (C) After resection of the necrotic round and falciform ligaments, extension of the preperitoneal fat necrosis over the diaphragm is visualized. Note patchy areas of subcutaneous fat necrosis. (D) Saponification fat necrosis involving greater omentum. Note patchy areas of subcutaneous fat necrosis (encircled).

The patient presented to our hospital sedated and mechanically ventilated, requiring 80% oxygen and a positive end-expiratory pressure 14 cm H₂O. At no time during the hospital course did the patient require vasopressors. WBC increased to 23 K/mm³, and other laboratory tests were comparable to ones from the outside hospital. Lipase normalized to 43 U/L. The patient was taken urgently for laparotomy. Frank necrosis of the falciform ligament and preperitoneal fat extending onto the diaphragm were encountered (Fig. 1B) and resected within the limits of viable tissues (Fig. 1C). Scattered small areas of subcutaneous fat necrosis were encountered during midline laparotomy (Fig. 1C). There were areas of saponification fat necrosis involving greater omentum (Fig. 1D) typical for pancreatitis, without peripancreatic or mesenteric necrosis. Subsequent cultures of the cloudy ascites and the necrotic fat were negative. Some fluid in the necrotic falciform ligament resembled bile. Cholangiography was performed through the fundus of the non-inflamed gallbladder to rule out bile leak and demonstrated patency of the cystic duct and normal filling of the biliary tree. Right chest tube was placed given preoperative pleural effusion on imaging with drainage of serous fluid. Given the extent of necrosis and concern that it may progress, the abdomen was left open and managed with vacuum closure dressing. Re-exploration was performed in 48 hours where no additional debridement was needed. Definitive abdominal closure was deferred.

Postoperatively the patient required paralysis for severe hypoxemic respiratory failure with PaO₂/FiO₂ ratio of 105. The patient underwent chest CT that demonstrated residual right pleural effusion and pericardial fat edema (Fig. 2A).

Figure 2.

(A) Chest CT demonstrating right pleural effusion and pericardial fat inflammation. (B) Right thoracoscopy with debridement of the necrotic pericardial fat.

On hospital day 5 after some improvement of the respiratory status, the abdomen was closed. Following abdominal closure after repositioning into decubitus position, the patient underwent right thoracoscopy for drainage of pleural effusion and assessment of the anterior mediastinum, given fever and ongoing leukocytosis. Necrotic pericardial fat was encountered and resected as much as was practical from a right-sided approach (Fig. 2B). Given acute respiratory distress syndrome (ARDS) and desaturations, only short periods of intermittent single lung ventilation were possible. Additional resection would require left-sided approach for complete debridement and visualization of the left phrenic nerve, but given difficulties of intraoperative ventilation, contralateral thoracoscopy was not performed immediately. By hospital day 9, the WBC decreased to 17, ARDS had resolved, and the patient was extubated.

On hospital day 11, the patient had an episode of supraventricular tachycardia and leukocytosis increased to 24. A chest CT showed resolution of the right pleural effusion and worsening edema of residual pericardial fat on the left (Fig. 3A). We considered these signs as an escalation of systemic inflammatory response requiring surgical debridement. Left thoracoscopy was performed with the finding of extensive necrosis of pericardial fat (Fig. 3B). Complete resection was performed between the diaphragm and greater vessels and extended across the midline into the right pleural space (Fig. 2B–D). A resected specimen confirmed extensive fat necrosis with no visualization of organisms (Fig. 4 A–B). All tissue and fluid cultures were negative. After complete VATS debridement, leukocytosis promptly resolved and no new tachyarrhythmia events were noted.

Figure 3.

(A) Chest CT demonstrating resolved right pleural effusion and worsening edema of the pericardial fat. (B) Left thoracoscopy with finding of pericardial fat necrosis. (C) Left thoracoscopy demonstrating resection of the necrotic pericardial fat. (D) Left thoracoscopic complete resection of the pericardial fat across the anterior mediastinum. A window on the parietal pleura overlying the contralateral right lung is resected as well (dotted line).

Figure 4.

(A) Resected pericardial fat during left thoracoscopy demonstrating extensive necrosis. (B) Histology of resected pericardial adipose tissue (100 × magnification). Note the fat necrosis characterized by loss of adipocyte nuclei, increased mixed inflammation, and degraded architecture present in a gradient of increasing severity toward the bottom of the image.

The patient was discharged on hospital day 27 after a slow but otherwise uncomplicated recovery. A chest CT in 1 month demonstrated complete resection of pericardial fat and absence of intrathoracic inflammation (Fig. 5). The patient is alive and well after 4 years with no subsequent hospitalizations or complaints.

Figure 5.

Chest CT at the time of the follow-up demonstrates complete resection of the pericardial fat.

Additional workup demonstrated hemoglobin A1C 5.1% and alpha-1 antitrypsin 318 mg/dL. Autoantibodies to neutrophil cytoplasmic antigens, anti-smooth muscle, antinuclear, and antistreptolysin O (ASO) antibodies titers were within reference range.

Discussion

Our patient presented with intraabdominal, pericardial, and subcutaneous non-infected fat necrosis. While the cause cannot be proven, based on imaging, intraoperative, and clinical findings, we consider chronic calcific pancreatitis in the setting of the hypermetabolic syndrome to be the etiology of disseminated fat necrosis.

Our patient had chronic calcific pancreatitis, likely tobacco-, alcohol-related, and confounded by a metabolic syndrome. Atrophy of the gland and pancreatic duct dilation indicate prior episodes of inflammation. Some patients, like ours, are newly diagnosed with chronic pancreatitis based on imaging findings without prior history of severe documented episodes of abdominal pain or acute pancreatitis. Acute abdominal pain was likely related to development of subcutaneous and abdominal fat necrosis with peritoneal irritation and a flare up of pancreatitis.

The patient did have a documented mild surge in lipase level above the reference range.

Enzymes were not elevated three times the upper limit of normal, a most commonly used criterion for diagnosis of acute pancreatitis, which may raise a question about association of the fat necrosis with pancreatitis. In clinical practice, however, even clinically severe pancreatitis may be associated with only minimal or no elevation in amylase and lipase^[12,13^]. In patients with chronic atrophic pancreatitis exocrine secretory function is frequently exhausted due to destruction of acinar sell. In some patients with recurrent attacks, peak levels of enzymes tend to decrease with each episode. Additionally, the plasma half-life of lipase is 7–14 hour and depends on etiology of pancreatitis^[14^]. This may result in the initial elevation of serum lipase being undetected with delayed presentation of the patient. Presence of the triglycerides with circulating inhibitors may result in normal ranges of serum enzymes as well.

No other etiologies of sterile intraabdominal saponification fat necrosis are seen in clinical practice other than pancreatitis. The urine test for trypsinogen-2 that has higher sensitivity and specificity for acute pancreatitis was not performed due to unavailability of the test during the hospital course.

Lack of pancreatic and peripancreatic inflammation on CT in this case is similarly related to severe atrophy of the gland. Even in acute episodes confirmed by laboratory tests, the CT imaging of acute interstitial and chronic exacerbation cases may have no intrinsic pancreatic and peripancreatic changes (e.g., Balthazar grade A or Modified CT severity index score 0–2).

We did not perform endoscopic retrograde cholangiopancreatography or magnetic resonance cholangiopancreatography, as there was no suggestion of fluid collections or disruption of dilated pancreatic duct on an abdominal CT scan or during abdominal exploration. The role of cholecystectomy in this case may be arguable. As the cholangiogram showed no gallbladder obstruction, and the gallbladder was normal in appearance, the decision was made to not perform cholecystectomy and focus instead on excising areas of fat necrosis. One could also argue that the cholangiogram was not needed and gallbladder emptying could be ascertained with compression of the gallbladder; however, some the fluid in the necrotic tissue resembled bile. Cholangiogram was indicated to rule out leak from the biliary system. Incidental cholecystectomy for a gallbladder without stones or pathology was not felt to be indicated.

All abdominal and chest intraoperative cultures had no growth, which is not uncommon in pancreatitis. Non-infected peripancreatic necrosis in the abdomen in the setting of pancreatitis is typically managed without debridement if a patient’s condition allows. This was the approach with pericardial fat necrosis we initially took in the setting of ongoing ARDS. Later, our patient had an exacerbation of the systemic inflammatory response manifested by a new surge in WBC and tachyarrhythmia. We then elected to maximally clear the mediastinum of the non-viable tissues with left thoracoscopy to assure safe recovery and prevent septic complications.

We were not able to find any similar reports of intraabdominal fat necrosis with diffuse involvement of the pericardial fat that were not associated with pancreatic necrosis, pseudocyst, or fistula formation. Additionally, our patient had focal panniculitis seen during laparotomy, not visualized on preoperative CT. This suggests that subclinical panniculitis may be more common in pancreatitis than reported and contribute to abdominal pain and tenderness.

Anecdotal cases of focal pericardial (epicardial) fat necrosis represent a completely separate entity. It is not associated with pancreatitis. Patients are typically middle-aged with moderate obesity and present with acute-onset lower chest pain. Presentation is accompanied with normal electrocardiogram and imaging demonstrating well-defined soft tissue density, usually in pericardiophrenic angle on the side of the chest pain. Resection is usually needed to provide diagnosis. Microscopy demonstrates hemorrhagic necrosis in early and dense fibrosis in chronic lesions, indicating a different pathophysiologic mechanism from pancreatic panniculitis and likely related to torsion of the fat lobule. Surgical options include open or thoracoscopic resection^[15^]. Observation is an accepted contemporary approach for minimally symptomatic patients.

Purulent and necrotizing mediastinitis secondary to esophageal perforation or descending oropharyngeal infections are common and are routinely managed thoracoscopically.

Based on a thorough search of the world literature, this is the first report of thoracoscopic resectional debridement of non-infected diffuse pericardial fat necrosis in the setting of pancreatitis in the absence of fistula or pseudocyst. We suggest that, in rare cases of mediastinal fat necrosis in the setting of pancreatitis, thoracoscopic debridement is technically feasible and may be considered as a management option if the patient manifests signs of ongoing systemic inflammatory response.

Ethical approval

This is a case report from the University of Iowa Hospitals and Clinics. It does not constitute research and does not require ethical approval.

Consent

Written informed consent was obtained from the patient for publication and any accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal on request.

Sources of funding

None.

Author contributions

M.H.: Writing the paper; D.S.: Writing the paper; A.S.: Data analysis and interpretation; F.J.: Data analysis and interpretation; M.T.: Data analysis and interpretation; E.V.A.: Writing the paper.

Conflicts of interest disclosure

The authors report no conflicts of interest.

Research registration unique identifying number (UIN)

Not applicable.

Guarantor

Evgeny V. Arshava.

Provenance and peer review

Not commissioned; externally peer-reviewed.

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Central picture legend

Intraoperative photograph demonstrating resection of the necrotic pericardial fat.

Central message

We present a case of concomitant non-infected diffuse pericardial, intraabdominal, and subcutaneous fat necrosis that required thoracoscopic and intraabdominal resection.