From the Archives: Corticosteroids and Prophylaxis. What complications should you try to prevent in patients on chronic corticosteroids?

December 7, 2017


Please enjoy this post from the archives dated October 30, 2013

By Robert Joseph Fakheri, MD

Faculty Peer Reviewed

A 55 year-old male is recently diagnosed with systemic sarcoidosis. The patient is started on prednisone 40mg with the plan to decrease the dose after remission of symptoms, which may take a number of months. What kind of prophylaxis should the patient receive?

Corticosteroids are an effective treatment option for a number of diseases spanning many specialties. However, long-term corticosteroid treatment is marred with a number of side effects including hypertension, hyperglycemia, weight gain, adrenal suppression, osteoporosis, peptic ulcer disease (PUD), and increased risk of infections [1,2]. In general, the risk of side effects has a direct relationship with dose and duration of treatment. In the past, the standard of care would be to monitor for these effects and address them accordingly. But with our expanding armamentarium of medications and growing literature, when is it appropriate to prevent complications before they even start? The most common targets of prophylaxis are the latter three: osteoporosis, PUD, and infections (specifically Pneumocystis jiroveci pneumonia or PCP).

Starting with osteoporosis, the agents available for prophylaxis are calcium, vitamin D, bisphosphonates, and lastly teriparatide (an analogue of parathyroid hormone). The literature has been reviewed by the American College of Rheumatology (ACR) and they issued guidelines with their recommendations [3]. Firstly, per their recommendations, all patients should be given 1,200-1,500mg of calcium per day and 800IU-1,000IU of vitamin D per day, or enough to achieve a therapeutic level of 25-hydroxyvitamin D. Though there is some variation based on age and gender, the general consensus is that everyone should have baseline dual-energy X-ray absorptiometry (DEXA) scan and that patients with high fracture risk (post-menopausal women, men older than 50 years old, low DEXA scores) on a dose equivalent of prednisone 7.5mg for more than 3 months should also be on a bisphosphonate. Very high-risk groups should be started on a bisphosphonate even with 5mg of prednisone for 1 month.

Though calcium and vitamin D are widely encouraged, the actual evidence is fairly limited. One study of 62 patients (on average prednisone dose from 16-21mg) compared the combination of vitamin D 50,000IU weekly and calcium 1,000mg daily to placebo and found no difference in loss of bone mineral density (BMD) after 3 years of follow-up [4]. A similar study of 17 patients with inflammatory bowel disease found no benefit in BMD after 1 year (calcium 1,000mg plus vitamin D 250IU daily with average prednisone dose 12-14mg) [5]. Another study of 41 women on average prednisone dose of 15mg/day compared 0.25-1ug/day of alfacalcidol (a vitamin D analogue) to 500mg/day of calcium over a period of 3 years and found that the former was able to maintain BMD while patients on calcium alone lost BMD as early as 6 months into therapy [6]. Lastly, a study of 81 patients with systemic lupus erythematosus (SLE) on average prednisone dose of 10-11mg compared calcium 1,200mg/day plus calcitriol 0.5ug/day against calcium 1,200mg/day alone against placebo found a minimal increase in the combination group, but not significantly different than the other two groups over 2 year study period [7]. It makes sense in this population to screen for vitamin D deficiency and treat when necessary, but routine supplementation has limited evidence. Although calcium and vitamin D are considered benign, they may be problematic in patients with hypercalcemia such as sarcoidosis or other granulomatous diseases and may interfere with absorption of other medications such as mycophenolate mofetil.

On the other hand, data supporting bisphosphonates is more robust. One of the first studies randomized 477 patients receiving corticosteroids (average prednisone dose of 9-10mg/day) with diagnoses across specialties over 48-weeks to either alendronate 5mg, alendronate 10mg or placebo [8]. All patients received 800-1,000mg of calcium and 250-500IU of vitamin D daily. The authors found that both of the bisphosphonate groups had increased BMD in lumbar spine (+2-3%) and femoral-neck (+1%), while the placebo group had decreased BMD in lumbar spine (-0.4%) and femoral neck (-1.2%), p<0.01.

Different specialty groups have individually confirmed this benefit of bisphosphonates. In rheumatology, a study of 200 patients with rheumatic diseases compared alendronate 10mg to alfacalcidol over 18 months and found an increase in BMD by +2.1% in bisphosphonate group compared to loss of BMD by -1.9% in alfacalcidol group, with a net difference of 4%[9]. In dermatology, 29 patients with immunobullous disease were randomized to alendronate or placebo for 12 months and found increases in BMD in the treatment group by +3.5% to +3.7% in lumbar spine and femoral neck respectively as opposed to decreased in BMD in the control group by -1.4% and -0.7% (p=0.01) [10]. In gastroenterology, a study of 39 patients with ulcerative colitis were randomized to receive alendronate 5mg or alfacalcidol for 12 months; the investigators demonstrated an increase in lumbar spine BMD in the bisphosphonate group by +4.1% compared with +0.9% in the alfacalcidol group (p<0.0005), though smaller differences were observed in femoral neck BMDs that did not meet statistical significance [11]. Lastly, in pulmonology, a study of 30 patients with sarcoidosis were randomized to either alendronate 5mg or placebo for 12 months and found a change in radial BMD of +0.8% in the treatment group compared with -4.5% in the placebo group (p<0.01)[12]. Although BMD is merely a surrogate marker for fracture risk with inconsistent correlation, it is commonly used due to its facile measurement and will likely remain the measuring stick until better data is available on fracture risk [13].

Moving on to PUD, there is a theoretical benefit of reducing gastric pH with agents such as proton-pump inhibitors (PPIs) to prevent steroid-induced peptic ulcers. However, there is limited data on the subject largely because, despite much notoriety, there is limited data to even show that corticosteroids actually cause peptic ulcers in the first place. Initial reports of an association date as far back as 1951 with case series and studies showing the effect of corticosteroids to increase gastric acidity [14]. Since then, a multitude of trials followed by meta-analyses have reviewed the topic. In 1976, a meta-analysis of over 3558 patients from 26 prospective, randomized, double-blind, placebo-controlled trials found no difference in ulcer risk over placebo with an absolute risk of about 1% [15]. Another meta-analysis in 1983 challenged these findings and found a relative risk (RR) of 2.3 (CI 1.4 – 3.7) [16]. However, this analysis was critiqued primarily for its use of non-blinded studies and suspected cases of PUD based on symptoms of severe dyspepsia [17].

Then, in 1991, a nested case-control study of 1415 patients found that ulcer risk was only increased in concurrent users of non-steroidal anti-inflammatory drugs (NSAIDs): RR 1.1 for corticosteroid alone, RR 4.4 for corticosteroid use in patients already on NSAIDs, RR 14.6 for corticosteroid and NSAIDs compared to controls [18]. This suggested that corticosteroids themselves do not cause ulcers by themselves, but may impair wound-healing to exacerbate ulcers caused by other etiologies. In 2001, a similar nested case-control study from the UK of 2105 cases looking at upper gastrointestinal complications found odds ratios (OR) of 1.8 for users of corticosteroids alone, 4.0 for users of NSAIDs alone, and 8.9 for users of both [19].

Thus, the debate continues. It is clear that corticosteroids in conjunction with NSAIDs is a major risk factor for PUD and should be avoided; when unavoidable, patients would likely benefit from acid suppressive therapy. In fact, use of NSAIDs alone should warrant consideration of PPI prophylaxis as some data shows they can reduce hospitalization for PUD complications from NSAID use by 67% [20]. Given the data that corticosteroids increase gastric acidity, this mechanism may contribute to symptoms of dyspepsia without mucosal breakdown; consequently, acid suppressive therapy may relieve these symptoms. Therefore, given the lack of evidence and multiple potential complications of PPIs including enteric infections like Clostridium difficile colitis, nutrient malabsorption, pneumonia, gastrointestinal neoplasms, acute interstitial nephritis, and gallbladder dyskinesia [21] , the role of PPIs should largely be reserved for treatment rather than prophylaxis of gastrointestinal complications of corticosteroids.

Lastly is the topic of PCP prophylaxis. While there are clear data and guidelines for patients with acquired immune deficiency syndrome (AIDS), the data is not as clear for other states of immunosuppresion. Although the first-line agent trimethoprim-sulfamethaxoazole (TMP-SMX) is very effective at preventing PCP, it comes with its own consequences including adverse drug reactions, cost, and risk of antibiotic resistance.

As before, the risk of PCP is correlated with dose and duration of corticosteroid use, but with PCP there is the added variable of the relative immune dysregulation caused by the underlying disease being treated. A retrospective study of 116 PCP patients without AIDS were analyzed and compared by underlying disease process including hematologic malignancy, solid tumors, organ transplants, inflammatory diseases, and other. With inflammatory diseases, the median dose of prednisone at time of diagnosis was 40mg (range 12-100, interquartile range 20-56) with a median duration of 16 weeks (range 5-672, interquartile range 8-60) [22]. Another retrospective study of 15 patients with SLE and PCP compared with matched controls found that patients that developed PCP had higher doses of prednisone (49 vs. 20mg), lower total lymphocyte counts (1040 vs. 1842 cells/mm3), and lower CD4 lymphocyte counts (156 vs. 276). The authors suggested prophylaxis when total lymphocyte count was less than 750 or CD4 count less than 200 [23].

So CD4 counts appear to be a useful tool in assessing risk, but other factors also contribute such as lung architecture. In a retrospective study of 74 patients with interstitial lung disease on corticosteroids, 7 patients developed PCP. The mean dose at time of diagnosis was prednisone 37mg with mean duration of 10 weeks. CD4 counts ranged from 59 to 836, with a mean of 370 [24]. The authors argued that due to their underlying lung disease, the patients were at higher risk for PCP and became infected at higher CD4 counts than patients with other underlying diseases.

In some cases, such as patients with bone marrow transplants, PCP prophylaxis is recommended regardless of corticosteroid use [25]. A meta-analysis of transplant patients and hematologic malignancies estimated the number needed to harm from prophylaxis as 32, and thereby recommended prophylaxis for patients with a risk of 3.5% or higher [26]. This risk is difficult to determine without good quality studies on specific patient populations, though clearly the risk will be modified by individual patient co-morbidities and concurrent therapies. Patients that meet this criterion include patients with transplants, acute lymphoblastic leukemia, severe combined immunodeficiency syndrome, and Wegener’s granulomatosis [26].

Using the above metric as a guide, routine prophylaxis is not indicated in the average patient with skin disease requiring immunosuppressive therapy. In one study of 198 dermatology patients on immunosuppressive therapy, only 0.7% of at-risk patients developed PCP. The majority of patients, 79%, were on corticosteroids (either alone or in conjunction with other agents) with median duration of 28.5 months (average dosage unavailable) [27].

Multiple authors have recommended prophylaxis in patients with underlying immunologic disorder or malignancy receiving equivalent prednisone daily dose of 20mg or more for at least 1 month [28,29], but the current data suggests that this one-size-fits-all approach may subject many patients unnecessarily to prophylaxis and also fail to protect a number of patients at significant risk for PCP. Although unfortunate, the decision for PCP prophylaxis ultimately rests on a clinician’s assessment of multiple variables including not only corticosteroid dose and duration, but also underlying disease and co-morbidities that may affect immune function and lung architecture, in addition to concurrent therapies and measurements of total lymphocyte counts and CD4 counts.

In summary, there is fairly strong evidence for DEXA screening and bisphosphonate use for osteoporosis prevention in most patients on chronic corticosteroids (particularly patients greater than age 50 on prednisone equivalent dose of 7.5mg or more for at least 3 months), hardly any evidence for acid suppressive therapy for PUD prevention, and the need for a case-by-case risk assessment for antibiotic use for PCP prevention. Despite this evidence, there appears to be a lack of awareness in the medical community. A study of 360 physicians in the Czech Republic (100 from Gastroenterology, 100 from General Practice, 80 from Pulmonology/Immunology, and 80 from Neurology/Neurosurgery) found that 82% of physicians (61% of gastroenterologists) believed that corticosteroids significantly increase the risk of PUD and 75% of physicians (55% of gastroenterologists) believed that gastroprotective therapy was appropriate for patients on systemic corticosteroids alone without concurrent therapy with NSAIDs [30]. Moreover, despite the evidence for initiating therapy with bisphosphonates, they are still not widely used. One study in the United Kingdom assessed adherence to guidelines by rheumatologists and found about half did not order a DEXA scan when it was indicated [31]. Another study in a multispecialty rheumatology urban practice found that only 39% patients with rheumatoid arthritis on chronic corticosteroids received the recommended DEXA screening and treatment/prophylaxis according to ACR guidelines. The literature in dermatology is similar with only 20% patients referred to a tertiary center on chronic oral corticosteroids for median duration of 6 months that had been received bisphosphonates [32]. The reason for this discrepancy between evidence and practice is unclear, but likely in part due to limited awareness of the scientific literature as demonstrated in the above survey. Moreover, treating physicians may have varying comfort levels in prescribing different medications (e.g. high comfort with PPIs leading to over-prescription and low comfort with bisphosphonates leading to under-prescription).Thus, across specialties, there is a need to increase awareness to increase the use of screening DEXA scans, increase the use of bisphosphonates, and decrease the use of acid-suppressive therapy for patients on chronic corticosteroids.

Dr.  Robert Joseph Fakheri is a 3rd year resident at NYU Langone Medical Center

Peer reviewed by Peter Izmirly, MD, Division of Rheumatology, NYU Langone Medical Center

Image courtesy of Wikimedia Commons


1. Buchman AL. Side effects of corticosteroid therapy. J Clin Gastroenterol. 2001;33(4):289-294.

2. Huscher D, Thiele K, Gromnica-Ihle E, et al. Dose-related patterns of glucocorticoid-induced side effects. Ann Rheum Dis. 2009;68(7):1119-1124.

3. Grossman JM, Gordon R, Ranganath VK, et al. American college of rheumatology 2010 recommendations for the prevention and treatment of glucocorticoid-induced osteoporosis. Arthritis Care Res (Hoboken). 2010;62(11):1515-1526.

4. Adachi JD, Bensen WG, Bianchi F, et al. Vitamin d and calcium in the prevention of corticosteroid induced osteoporosis: A 3 year followup. J Rheumatol. 1996;23(6):995-1000.

5. Bernstein CN, Seeger LL, Anton PA, et al. A randomized, placebo-controlled trial of calcium supplementation for decreased bone density in corticosteroid-using patients with inflammatory bowel disease: A pilot study. Aliment Pharmacol Ther. 1996;10(5):777-786.

6. Lakatos P, Nagy Z, Kiss L, et al. Prevention of corticosteroid-induced osteoporosis by alfacalcidol. Z Rheumatol. 2000;59 Suppl 1:48-52.

7. Lambrinoudaki I, Chan DT, Lau CS, Wong RW, Yeung SS, Kung AW. Effect of calcitriol on bone mineral density in premenopausal chinese women taking chronic steroid therapy. A randomized, double blind, placebo controlled study. J Rheumatol. 2000;27(7):1759-1765.

8. Saag KG, Emkey R, Schnitzer TJ, et al. Alendronate for the prevention and treatment of glucocorticoid-induced osteoporosis. Glucocorticoid-induced osteoporosis intervention study group. N Engl J Med. 1998;339(5):292-299.

9. de Nijs RN, Jacobs JW, Lems WF, et al. Alendronate or alfacalcidol in glucocorticoid-induced osteoporosis. N Engl J Med. 2006;355(7):675-684.

10. Tee SI, Yosipovitch G, Chan YC, et al. Prevention of glucocorticoid-induced osteoporosis in immunobullous diseases with alendronate: A randomized, double-blind, placebo-controlled study. Arch Dermatol. 2012;148(3):307-314.

11. Kitazaki S, Mitsuyama K, Masuda J, et al. Clinical trial: Comparison of alendronate and alfacalcidol in glucocorticoid-associated osteoporosis in patients with ulcerative colitis. Aliment Pharmacol Ther. 2009;29(4):424-430.

12. Gonnelli S, Rottoli P, Cepollaro C, et al. Prevention of corticosteroid-induced osteoporosis with alendronate in sarcoid patients. Calcif Tissue Int. 1997;61(5):382-385.

13. Divittorio G, Jackson KL, Chindalore VL, Welker W, Walker JB. Examining the relationship between bone mineral density and fracture risk reduction during pharmacologic treatment of osteoporosis. Pharmacotherapy. 2006;26(1):104-114.

14. Gray SJ, Benson JA, Jr., Reifenstein RW. Chronic stress and peptic ulcer. I. Effect of corticotropin (acth) and cortisone on gastric secretion. J Am Med Assoc. 1951;147(16):1529-1537.

15. Conn HO, Blitzer BL. Nonassociation of adrenocorticosteroid therapy and peptic ulcer. N Engl J Med. 1976;294(9):473-479.

16. Messer J, Reitman D, Sacks HS, Smith H, Jr., Chalmers TC. Association of adrenocorticosteroid therapy and peptic-ulcer disease. N Engl J Med. 1983;309(1):21-24.

17. Conn HO, Poynard T. Adrenocorticosteroid therapy and peptic-ulcer disease. N Engl J Med. 1984;310(3):201-202.

18. Piper JM, Ray WA, Daugherty JR, Griffin MR. Corticosteroid use and peptic ulcer disease: Role of nonsteroidal anti-inflammatory drugs. Ann Intern Med. 1991;114(9):735-740.

19. Hernandez-Diaz S, Rodriguez LA. Steroids and risk of upper gastrointestinal complications. Am J Epidemiol. 2001;153(11):1089-1093.

20. Vonkeman HE, Fernandes RW, van der Palen J, van Roon EN, van de Laar MA. Proton-pump inhibitors are associated with a reduced risk for bleeding and perforated gastroduodenal ulcers attributable to non-steroidal anti-inflammatory drugs: A nested case-control study. Arthritis Res Ther. 2007;9(3):R52.

21. Cote GA, Howden CW. Potential adverse effects of proton pump inhibitors. Curr Gastroenterol Rep. 2008;10(3):208-214.

22. Yale SH, Limper AH. Pneumocystis carinii pneumonia in patients without acquired immunodeficiency syndrome: Associated illness and prior corticosteroid therapy. Mayo Clin Proc. 1996;71(1):5-13.

23. Lertnawapan R, Totemchokchyakarn K, Nantiruj K, Janwityanujit S. Risk factors of pneumocystis jeroveci pneumonia in patients with systemic lupus erythematosus. Rheumatol Int. 2009;29(5):491-496.

24. Enomoto T, Azuma A, Matsumoto A, et al. Preventive effect of sulfamethoxasole-trimethoprim on pneumocystis jiroveci pneumonia in patients with interstitial pneumonia. Intern Med. 2008;47(1):15-20.

25. Dykewicz CA. Summary of the guidelines for preventing opportunistic infections among hematopoietic stem cell transplant recipients. Clin Infect Dis. 2001;33(2):139-144.

26. Green H, Paul M, Vidal L, Leibovici L. Prophylaxis of pneumocystis pneumonia in immunocompromised non-hiv-infected patients: Systematic review and meta-analysis of randomized controlled trials. Mayo Clin Proc. 2007;82(9):1052-1059.

27. Lehman JS, Kalaaji AN. Role of primary prophylaxis for pneumocystis pneumonia in patients treated with systemic corticosteroids or other immunosuppressive agents for immune-mediated dermatologic conditions. J Am Acad Dermatol. 2010;63(5):815-823.

28. Sepkowitz KA. Pneumocystis carinii pneumonia without acquired immunodeficiency syndrome: Who should receive prophylaxis? Mayo Clin Proc. 1996;71(1):102-103.

29. Worth LJ, Dooley MJ, Seymour JF, Mileshkin L, Slavin MA, Thursky KA. An analysis of the utilisation of chemoprophylaxis against pneumocystis jirovecii pneumonia in patients with malignancy receiving corticosteroid therapy at a cancer hospital. Br J Cancer. 2005;92(5):867-872.

30. Martinek J, Hlavova K, Zavada F, et al. “A surviving myth”–corticosteroids are still considered ulcerogenic by a majority of physicians. Scand J Gastroenterol. 2010;45(10):1156-1161.

31. Wall E, Walker-Bone K. Use of bisphosphonates and dual-energy x-ray absorptiometry scans in the prevention and treatment of glucocorticoid-induced osteoporosis in rheumatology. Qjm. 2008;101(4):317-323.

32. Liu RH, Albrecht J, Werth VP. Cross-sectional study of bisphosphonate use in dermatology patients receiving long-term oral corticosteroid therapy. Arch Dermatol. 2006;142(1):37-41.