A Sausage-Shaped Pancreas and Poorly Controlled Diabetes: Rethinking First-Line Steroids
Steroids are the guideline-favored first move in treatment-naive IgG4-related disease — but the guideline wasn’t written for a patient whose diabetes is already this hard to control.
Painless jaundice and dark urine that his wife noticed before he did brought Frank D., a 61-year-old retired machinist, to the emergency department, where imaging showed diffuse, symmetric enlargement of his pancreas — the “sausage-shaped” appearance classically described in autoimmune pancreatitis — along with a stricture of his distal common bile duct. An urgent ERCP relieved the obstruction with a temporary stent, and biopsies taken during the same procedure, along with a markedly elevated serum IgG4 level, confirmed the diagnosis: dense lymphoplasmacytic infiltrate rich in IgG4-positive plasma cells, storiform fibrosis, and obliterative phlebitis — the pathologic triad that satisfies the 2020 revised comprehensive diagnostic criteria, the Japanese criteria set that admits a definite diagnosis on histology, rather than the 2019 ACR/EULAR classification criteria published the same year, which are a points-based research instrument. It is the triad that distinguishes IgG4-related disease from ordinary chronic pancreatitis or a malignant stricture, both of which were the real diagnoses this workup had to exclude first given how closely this presentation can mimic pancreatic cancer.
His type 2 diabetes, diagnosed eight years ago, has been genuinely difficult to control despite metformin and basal insulin — his most recent A1c came back at 9.8%, and he has had two emergency visits in the past year for hyperglycemic symptoms severe enough to need same-day insulin adjustment. That history is not incidental to today's decision: high-dose glucocorticoids, the standard first-line induction for this disease, are well known to worsen glycemic control substantially, and in a patient already struggling at this baseline, that is not a theoretical side effect to mention in passing but a near-certain complication the team has direct prior evidence he is prone to. The stent has already relieved the obstruction, so nothing forces the decision tonight. What is easy to miss is what the standard course actually commits him to: several weeks of high-dose glucocorticoid in a man who has twice this year needed same-day insulin adjustment at an A1c of 9.8%. The foreseeable endpoint is not a nuisance rise in his glucose. It is a third emergency visit, arriving in the weeks he is least able to absorb one.
Whose guideline default this actually is
For treatment-naive IgG4-related disease, glucocorticoids remain the guideline-recommended first move, and the response rate is genuinely strong — most patients improve rapidly and dramatically. I'd start there before reaching for a B-cell-depleting agent as a first exposure, given the infection risk and cost that come with rituximab.
I'd push back specifically on applying that default to him. An A1c of 9.8% with two hyperglycemic ED visits this year isn't a hypothetical steroid side effect — it's a near-certain complication given his own documented history. Carruthers and colleagues' prospective open-label trial — thirty patients, no control arm, but response in nearly all of them — showed real induction-phase remission with rituximab in this exact disease.
The guideline default assumes an average patient's cost from steroids. His diabetes means that assumption doesn't hold for him specifically, and I don't think we should treat a population-level recommendation as though it were individualized advice for this particular patient.
I agree his diabetes is a real cost, but I don't think either of you has actually quantified what a shorter, lower-dose steroid course with proactive glucose monitoring would cost him, versus assuming full-dose steroids at face value. Steroid-related hyperglycemia is often manageable with aggressive insulin adjustment rather than uniformly severe.
I'd propose combining a shorter, lower-dose steroid taper with rituximab from the start, alongside close glycemic monitoring on my end — capturing most of the rapid symptomatic benefit steroids offer while limiting his specific exposure, rather than choosing one drug class and fully excluding the other.
Rituximab induction was started alongside a shortened, lower-dose prednisone taper, with endocrinology following his glucose closely through the steroid course rather than waiting for a hyperglycemic complication to prompt reactive adjustment.
Agreed by all three voices once the combined plan was on the table: this represented a genuine third option rather than a compromise that satisfied no one — the Gastroenterologist accepted that a modified, shorter steroid course still honored the evidence for glucocorticoid response, while the Rheumatologist accepted that a reduced-dose course, closely monitored, meaningfully changed the specific diabetes risk that had motivated proposing rituximab alone in the first place.