Medicine Correspondence Blog
The Medicine Correspondence Blog allows authors to post Letters to the Editors, Reviews, and other editorial writings that are not considered original research.

Thursday, April 13, 2017

We thank Dr Morrison and CHIMERIX Laboratory for their interest in these two case reports. Here we bring some complementary data to improve the understanding of the manuscript and to answer to the critical points that were highlighted by the letter from Dr Morrison.

In the first case, prolonged CMV infection may be not an explanation for deterioration of renal function. First, prolonged CMV infection leads to a progressive and long-term decline of renal function(1). In this case, decrease of GFR occurs rapidly (few days) after initiation of either FCV and BCV, suggesting acute tubular injury. Second, no recovery is observed after CMV-induced decline of renal function(2). In this case, patient 1 showed two reversible acute renal injury after treatment, followed by an early improvement of renal function after discontinuation of FCV (Creatinine: 42 mg/L à 37 mg/L) and after discontinuation of BCV (58 mg/L à 24 mg/L after 4 weeks), compatible with the diagnosis of acute tubular injury. The estimated GFR also returned to baseline.  Moreover, after 2 years of follow-up, patient 1 does not show any decline of renal function (creatinine level at 31 mg/L in march 2017). But, according to the data, a potentiation of BCV-induced toxicity by FCV may not be excluded. Finally, observation of renal graft biopsy showed only an extended acute tubular injury(2,3). As previously published, prolonged CMV infection leads to long-term graft loss (3 years)(1). Two events are commonly observed: involvement of acute rejection, (Patient 1 did not present any graft rejection, even after 2 years), or observation of irreversible graft fibrosis(2,3) (high grade tubular atrophy and interstitial fibrosis) that were not observed in our biopsy. Moreover, as previously seen, in this case, acute renal injury was reversible.

As shown in Figure 1C, patient 2 was treated with FCV two times. At each time, patient 2 presented a FCV-mediated acute renal injury (creatinin level raised to 35 mg/L) which was reversible after discontinuation and rehydration (baseline creatinine level raised to 20-23 mg/L). Just after renal recovery of first FCV treatment, patient received BCV for persistent CMV viremia. Creatinin level raised from 22 mg/L to 30 mg/L (from 01/09/15 to 10/09/15) and BCV was discontinued. Although BCV was stopped, creatinine level continues to increase to 52 mg/L in a context of diarrhea and increase absorption of tacrolimus. In this context, patient presented concomitantly an overdosage of everolimus and tacrolimus responsible of secondary thrombotic micro-angiopathy (TMA). Although creatinine level increased after first introduction of BCV, in this context of drug-induced TMA, the responsibility of BCV cannot be clearly stated. Moreover, renal function was aggravated by a second FCV treatment (27/09/15).

After complete recovery of the episode (17/12/15), creatinin level decreased close to baseline (22-25 mg/L). A second BCV treatment for persistent CMV viremia led to a progressive increase of creatinine level (44 mg/L). Thus, BCV was discontinued in February 2016, also 4 months after last treatment using FCV and 3 months after recovery of the first episode. The kidney biopsy showed only extended tubular necrosis, that cannot be attributed to FCV treatment (prescribed 4 months before, in September 2015). Moreover, no TMA lesions were observed in the biopsy. Thus, during the reintroduction of BCV, at a distance from last FCV-mediated renal injury and TMA episode, without clinical event or treatment modification and without sequels of TMA in the biopsy, renal function declined. Discontinuation of BCV led to rapid improvement of renal function. We share the opinion that concomitant multidrug administration and nephrotoxic therapy (calcineurin inhibitor, foscavir) may contribute to increase the BCV-mediated renal toxicity.

The underlying mechanisms of BCV-mediated renal toxicity are unknown. As highlighted in the manuscript, BCV in not a substrate of OAT1. But several other transporters may be involved in BCV-mediated tubular toxicity such as OCT2, OCTN2, MRP2, MDR1, BCRP, MATE1, MATE2-K. Moreover, the presence of chronic kidney failure in patient may potent tubular toxicity(4).

Safety studies and clinical assays in healthy volunteers or in hematopoietic stem cells receivers assessing toxicity of BCV are reassuring. Indeed, patients showed great tolerance to BCV, particularly concerning the involvement of acute kidney injury. However, the previous published trials may not be representative of our two specific patients (solid organ transplant with previous severe chronic kidney failure and concomitant treatment with other nephrotoxic drugs):

Indeed, during CMX001-201 study(5), focusing on the prevention of CMV infection in hematopoietic stem cell receivers, patients with severe chronic kidney failure (eGFR < 30 mL/min) were excluded. Moreover, depending of the weekly dose, 40 to 78% of patients presented an elevation of creatinine level. Interestingly, the most common side-effect was diarrhea.

In CMX001-202 trial, realized in adults and pediatric population for the preemptive treatment of adenovirus in hematopoietic stem cell receiver, GFR analysis after treatment was not performed in adult subjects due to small number of patient(6). The same authors reviewed renal function from the 3 completed clinical studies using BCV for treatment of adenovirus infection. In total 72 kidney-related adverse events were reported in 62 patients who received BCV (total 420 patients). About 50% of renal adverse events were considered as moderate to severe. Five (7%) renal events led to discontinuation of BCV. Four (6%) renal failure events were considered possibly related to BCV by the reporting investigator(6).

In phase 3 studies, renal transplant receivers are poorly represented. In CMX001-304 study, an open-labeled phase 3 study to evaluate BCV in treatment of adenovirus infection, only one adult renal transplant receiver was recruited(7). In CMX001-350 study, an open-labeled trial to assess BCV efficiency during life threatening disease caused by dsDNA viruses, only 16% of patients receive a solid organ transplant without precision of organ (liver, renal, lung, heart or combine…) (7,8).

In summary, in previous published trials, BCV showed good renal tolerance but renal adverse events were already report in some patients and were considered as possibly related to BCV. Moreover, immunocompromised patients enrolled during phase 2 and 3 trials were mainly hematopoietic stem cell receivers without moderate or severe chronic kidney disease and may not be compared to our two patients. Thus, the concomitant presence of kidney allograft in adults with chronic kidney failure may contribute to BCV-mediated toxicity.

Concerning renal adaptation, we followed the recommendations of CHIMERIX laboratories, advising discontinuation of BCV when estimated GFR decreases under 15 mL/min/1,73 m2. For both patient, elevation of creatinine was followed by discontinuation of BCV when eGFR decrease under 15 mL/min/1,73 m2. But French national society of Nephrology and Dialysis edited evidence-based guidelines for renal adaptation of BCV in chronic kidney failure patients. We have to bring a correction in the manuscript (Paragraph 1.1 line 23-24, "In accordance with CHIMERIX" should be replaced by: " In accordance with French Society of Nephrology and Dialysis").  However, even with renal adapted doses, a renal toxicity was observed in both patients and for both patient, discontinuation of BCV led to improve renal function.  


1. Arthurs SK, Eid AJ, Pedersen RA, Kremers WK, Cosio FG, Patel R, et al. Delayed-Onset Primary Cytomegalovirus Disease and the Risk of Allograft Failure and Mortality after Kidney Transplantation. Clinical Infectious Diseases. 2008 Mar 15;46(6):840–6.

2. Freeman RB Jr. The "Indirect" Effects of Cytomegalovirus Infection. American Journal of Transplantation. 2009 Nov;9(11):2453–8.

3. Reischig T, Jindra P, Hes O, Bouda M, Kormunda S, Třeška V. Effect of Cytomegalovirus Viremia on Subclinical Rejection or Interstitial Fibrosis and Tubular Atrophy in Protocol Biopsy at 3 Months in Renal Allograft Recipients Managed by Preemptive Therapy or Antiviral Prophylaxis. Transplantation. 2009 Feb;87(3):436–44.

4. George B, You D, Joy MS, Aleksunes LM. Xenobiotic transporters and kidney injury. Advanced Drug Delivery Reviews. Elsevier B.V; 2017 Jan 19;:1–66.

5. Marty FM, Winston DJ, Rowley SD, Vance E, Papanicolaou GA, Mullane KM, et al. CMX001 to prevent cytomegalovirus disease in hematopoietic-cell transplantation. N Engl J Med. 2013 Sep 26;369(13):1227–36.

6. Tippin TK, Morrison ME, Brundage TM, Mommeja-Marin H. Brincidofovir Is Not a Substrate for the Human Organic Anion Transporter 1: A Mechanistic Explanation for the Lack of Nephrotoxicity Observed in Clinical Studies. Ther Drug Monit. 2016 Dec;38(6):777–86.

7. Grimley M, Florescu D, Danziger-Isakov L, Brundage T. Treatment of Adenovirus Infection in Solid Organ Transplant with Brincidofovir. chimerixcom

8. Chittick G. Preliminary Safety Results and Antiviral Activity from the Open-label Pilot Portion of a Phase 3 Study to Evaluate Brincidofovir (BCV) for the Treatment of Adenovirus (AdV) Infection. 2014 Nov 10;:1–19. 

Monday, March 6, 2017

Dear Editor, the recent publication on “dengue-associated hemophagocytic lymphohistiocytosis” is very interesting [1]. Chung et al. noted that “it would be important to differentially diagnose dengue-associated HLH from severe DHF [1].” I would like to share ideas and experience on this complication of dengue. In fact, the hemophagocyticlymphohistiocytosis is an important uncommon complication of dengue. In endemic area, the regular blood examination in follow-up dengue patient is recommended and it can be the way for early diagnosis of hemophagocytic lymphohistiocytosis [2]. The major pitfall in management of the case is the neglecting on the blood examination follow - up of the dengue patient [3]. Chung et al. might not recognize the condition at first, hence, they might not observe the problem at the early stage. Indeed, the useful tests in following up hospitalized dengue patients include complete blood count with blood smear examination, coagulation study and liver function test [2]. The observation of anemia and abnormal liver function during the course of dengue illness is the main finding indicating the risk of hemophagocytic lymphohistiocytosis in dengue patients [4].

Professor Viroj Wiwanitkit
Perofessor, senior expert, Surindra Rajabhat University, Thailand; Visiitng professor, Hainan Medical University, China; honorary professor, Dr DY Patil University, India

1. Chung SM, Song JY, Kim W, Choi MJ, Jeon JH, Kang S, Jung E, Noh JY, Cheong HJ, Kim WJ. Dengue-associated hemophagocyticlymphohistiocytosis in an adult: A case report and literature review.  Medicine (Baltimore). 2017 Feb;96(8):e6159. doi: 10.1097/MD.0000000000006159.
2. Wiwanitkit V.  Dengue fever: diagnosis and treatment. Expert Rev Anti Infect Ther. 2010 Jul;8(7):841-5.
3. Joob B, Wiwanitkit V.  Hemophagocytic lymphohistiocytosis syndrome in dengue hemorrhagic Fever: correspondence. Indian J Pediatr. 2015 Jun;82(6):575.
4. Ellis EM, Sharp TM, Pérez-Padilla J, González L, Poole-Smith BK, Lebo E, Baker C, Delorey MJ, Torres-Velasquez B, Ochoa E, Rivera-Garcia B, Díaz-Pinto H, Clavell L, Puig-Ramos A, Janka GE, Tomashek KM.  Incidence and Risk Factors for Developing Dengue-Associated Hemophagocytic Lymphohistiocytosis in Puerto Rico, 2008 - 2013. PLoS Negl Trop Dis. 2016 Aug 24;10(8):e0004939.

Friday, February 17, 2017

We read with interest the recent manuscript by Faure et al describing reversible acute kidney dysfunction in two solid-organ transplant patients treated with brincidofovir (BCV) for cytomegalovirus infection1, and would like to highlight an alternative interpretation of these cases. 

The two patients described had severe pre-existing kidney disease.  Prior to initiation of BCV, both had clear renal injury due to foscarnet toxicity.  In the first case, estimated GFR declined from 20 mL/min to ~ 12 mL/min after 4 weeks of BCV; the progressive kidney dysfunction and biopsy findings that were attributed to BCV may represent renal injury from the prolonged CMV infection2,3,4 and/or continued deterioration from foscarnet-induced toxicity5.  In the second case, two courses of foscarnet each led to kidney injury; and renal function was further compromised by a tacrolimus and everolimus overdose with associated thrombotic microangiopathy.  Though temporary improvement in creatinine following withdrawal of BCV was observed, progressive renal deterioration thereafter is suggestive of alternate etiologies.

Multiple simultaneous renal insults affect kidney function in different ways temporally.  While vasoconstrictive mechanisms due to elevated calcineurin levels can resolve within days of lowering doses, thrombotic microangiopathy can have a prolonged and unpredictable course.  These overlapping etiologies combined with variations in creatinine due to volume changes can make attribution of any small change in kidney function to one of multiple causes difficult.  

After oral BCV administration, peak plasma cidofovir concentrations are <1% of those observed after IV cidofovir administration6.  That, and the fact that BCV is not a substrate for human OAT1 greatly reduces the risk of cidofovir-like nephrotoxicity6.  BCV's reduced nephrotoxicity is supported by a safety database of >1200 BCV recipients, including Phase 1 healthy volunteers as well as Phase 2/3 subjects who are immunocompromised and virally infected, have complex courses, and take multiple medications, including nephrotoxic agents.  In an analysis of patients who switched from IV cidofovir or foscarnet to BCV for the treatment of viral infections, improvements in estimated GFR were observed in 83% of 66 subjects6.  Because recovery from prior insults is not universal (and because CDV is a metabolite of BCV), Chimerix continues to encourage close monitoring of renal indices and hydration status during BCV treatment.

An additional correction to the manuscript bears mention.  Because circulating CDV levels remain low even in the presence of severe renal insufficiency, Chimerix has never recommended dose adjustment of BCV for compromised renal function.  Rather, renal replacement therapy should be started when GFR declines to <15 mL/min to prevent accumulation of CDV; the recommended dose of BCV remains unchanged.

Corresponding author:

Marion E. Morrison, MD

Executive Medical Director

Chimerix, Inc.



1. Faure E, Galperine T, Cannesson O, et al. Case report: Brincidofovir-induced reversible severe acute kidney injury in 2 solid-organ transplant for treatment of cytomegalovirus infection.  Medicine (Baltimore). 2016;95(44):e5226.

2. Arthurs SK, Eid AJ, Pedersen RA, et al. Delayed-onset primary cytomegalovirus disease and the risk of allograft failure and mortality after kidney transplantation.  Clin Infect Dis 2008; 46:840-846.

3. Freeman RB Jr. The 'indirect' effects of cytomegalovirus infection.  Am J Transplant 2009; 9(11): 2453-2458.

 4. Reischig T, Jindra P, Hes O, et al; Effect of cytomegalovirus viremia on subclinical rejection or interstitial fibrosis and tubular atrophy in protocol biopsy at 3 months in renal allograft recipients managed by preemptive therapy or antiviral prophylaxis. Transplantation 2009;87(3):436-444.

5. Nyberg G, Svalander C, Blohme I, Persson H.  Tubulointerstitial nephritis caused by the antiviral agent foscarnet.  Transplant Int 1989;2(1):223-227.

 6. Tippin T, Morrison M, Brundage T, Mommeja-Marin H; Brincidofovir is not a substrate for the Human Organic Anion Transporter 1 (OAT1):  a mechanistic explanation for the lack of nephrotoxicity observed in clinical studies.  Therapeutic Drug Monitoring 2016;38(6):777-786.

The Author Reply can be found on the Author Correspondence Blog. 

Friday, February 17, 2017

We have seen the article by Segal and colleagues, who showed that higher elevation in SBP from admission to discharge is associated with greater mortality [1].  While we commend the authors, we also question the prognostic worth of SBP change.  The authors document that low admission SBP may portend poor outcomes.  Moreover, our team has recently introduced the SBP/RAP ratio as a prognostic factor in acute systolic heart failure [2].  One may thus ask: was the association between elevation in SBP and mortality discovered by Segal and colleagues partly induced by their mutual relationship with low admission SBP?  Indeed, the authors questioned whether those with greater SBP increases might collectively be more fragile.  Those in the upper quartile of SBP change had more frequent anemia and furosemide use as well as less favorable NYHA classifications.  The authors' subgroup analyses found that, both among those with admission SBP<140 (Figure 3A) and among those with admission SBP>140 (Figure 3B), there was greater mortality risk associated with the upper quartile of SBP change.


Even among those with admission SBP<140, there is heterogeneity in the absolute values of SBP.  A patient with an admission SBP of 100 is included, as is one with SBP=130.  However, with "more room to go up", the former patient seems more likely to experience a greater SBP increase.  Technically, we can cite the phenomenon of regression to the mean (RTM), the tendency of a variable that is extreme on one occasion to become less extreme upon subsequent measurement [3].  The RTM phenomenon is especially concerning when the variable analyzed exhibits considerable intra-patient fluctuation, of which an excellent example is SBP [4].  Thus, if a patient with an admission SBP of 100 is at inherently greater risk than one with SBP=130, then patients with larger SBP changes may also be at greater risk than ones with smaller changes.  For those with admission SBP>140, low admission SBP is not a factor, but persons with greater SBP change may more consistently experience hypertension than those with lesser SBP change.  That mortality risk may be greater for those with more consistent hypertension seems plausible.


Considering the bottom three quartiles of the full cohort, the use of SBP change from admission to discharge (rather than absolute SBP values) combined many patients in whom admission SBP was elevated (and therefore a target for therapeutic intervention) with some patients in whom admission SBP was normal, as the median and IQR were 150 and 130-170 respectively.  Hence, reductions in SBP in the first and second quartiles may have resulted at least partly from antihypertensive therapy, of which escalation may have been warranted when SBP>140.  Although the bottom quartiles did not differ significantly from the top quartile on most medications, the authors acknowledged the limitation of being unable to account for dose changes.  In summary, when the intensity of therapeutic interventions with the goal of lowering SBP is not examined, and considering the heterogeneity in absolute values of SBP, the prognostic worth of SBP change is questionable.


We thank the authors for their contribution and the editor for the opportunity to respond.  We also hope that calling attention to the RTM phenomenon has been useful, as other cardiovascular research studies may also depend on repeated measurements which are prone to marked intra-patient variability.   ​

Hesham R. Omar [1], Richard Charnigo [2], Maya Guglin [3]

​1. Internal Medicine Department, Mercy Medical Center, Clinton, Iowa

2. Department of Biostatistics, University of Kentucky, Lexington, KY

3. Division of Cardiovascular Medicine, Linda and Jack Gill Heart Institute, University of Kentucky, Lexington, KY


*Corresponding author

Hesham R. Omar

Internal Medicine Department

Mercy Medical Center, Clinton, Iowa, USA

Cell: 312-714-9272


Dr. Charnigo has been a co-investigator on two grants from AstraZeneca and has traded IBM stock and/or stock options.


1. Segal O, Segal G, Leibowitz A, Goldenberg I, Grossman E, Klempfner R. Elevation in systolic blood pressure during heart failure hospitalization is associated with increased short and long-term mortality. Medicine (Baltimore). 2017 Feb;96(5):e5890.

2. Omar HR, Charnigo R, Guglin M. Ratio of systolic blood pressure to right atrial pressure is a novel marker to predict morbidity and mortality in acute systolic heart failure. American journal of cardiology. In press

3. Barnett AG, van der Pols JC, Dobson AJ. Regression to the mean: what it is and how to deal

with it. Int J Epidemiol 2005;34:215–20.

4. Bland JM, Altman DG. Some examples of regression towards the mean. BMJ 1994;309:780.

Thursday, January 12, 2017

We read with interest the paper by Ohira et al. of thyroid ultrasound examinations in Fukushima, which examines the relation between external radiation dose and thyroid cancer prevalence among Fukushima children.1 However, we point out that their classification of 59 municipalities in Fukushima prefecture into 3 areas is inappropriate. The "lowest dose area" was constituted of Aizu area with least thyroid dose and a distant Iwaki city with the highest thyroid dose, which led to a wrong conclusion that the external radiation dose was not associated with thyroid cancer prevalence among Fukushima children.  

​Ohira et al. of Fukushima Medical University examined the association between the prevalence of thyroid cancer and radiation dose among Fukushima residents.1 They used external radiation dose estimated by Fukushima Health Management Survey (FHMS) based on individual external doses from behavior data of 26.4% residents who responded the questionnaire.2 They classified municipalities based on fraction of respondents: "highest dose area" (≥1% received external radiation exposure of ≥5 mSv), "lowest dose area" (≤1% received ≥1 mSv), and the other "middle dose area". Mainly because the prevalence of thyroid cancer was found not to decrease in this order of decreasing external dose, they concluded that external dose due to nuclear accident is not associated with thyroid cancer prevalence. However, their classification of municipalities based on fraction of residents (1%) whose exposure exceeds 5mSv and 1mSv does not represent the average exposure dose in each municipality. Moreover, they seem to have made a serious mistake in their classification as follows.

Internal exposure to I-131 is known to be closely associated with the incidence of thyroid cancer among children. The UNSCEAR report on absorbed dose in thyroid 3a,b shows that the Ohira et al.'s "lowest external dose area" is composed of the Aizu area with least thyroid dose and a distant Iwaki city with the highest thyroid dose in Fikushima prefecture except evacuation zone. In a recent estimation of internal thyroid dose using a combination of thyroid measurement data, whole-body counter measurement data and atmospheric transport dispersion simulations, the residents of three municipalities including Iwaki city were shown to have the highest thyroid equivalent dose in Fukushima prefecture. 4 The "lowest external dose area" is found to be composed of the lowest dose Aizu area and the highest thyroid dose Iwaki city. Incidence rates of thyroid cancer for the highest, middle and lowest external dose areas in ref. 1 and those for Aizu and Iwaki districts constituting the "lowest dose area" are listed in Table 1. The average total effective dose to 10-year-old children estimated by UNSCEAR3b,c shows that the effective dose of Iwaki city is the 18th highest in 59 municipalities. If Iwaki city is classified as "middle external dose area" instead of "the lowest external dose area", dose response of thyroid cancer prevalence may be recovered.

​The conclusion of ref. 1 that external radiation dose due to nuclear accident is not associated with thyroid cancer prevalence among Fukushima children is found to come from the wrong constitution of the "lowest dose area" as a sum of Aizu with lowest thyroid dose and Iwaki city with the highest thyroid dose. External radiation dose estimation may possibly reveal the dose dependence of thyroid cancer, if it is used carefully with referring to various estimations of external dose and absorbed dose in thyroid. 


Corresponding author

Toshiko Kato, Dr. Science (Kyoto University)

11-3-604 Kangoku-cho, Nara, 630-8242, Japan


1. Ohira, T, Takahashi H, Yasumura S, et al.  Comparison of childhood thyroid cancer prevalence among 3 areas based on external radiation dose after the Fukushima Daiichi nuclear power plant accident: The Fukushima health management survey. Medicine. 2016; 95(35): p e4472 doi: 10.1097/MD.0000000000004472

2. Fukushima Health Management Survey, Basic Survey Appendix: Estimated external radiation dose, Web site.  Published Sep. 15, 2016, Accessed January 11, 2017.

3. UNSCEAR 2013 Report Vol. I Sources, Effects and Risks of Ionizing Radiation.  Published 2014 March, Accessed January 11, 2017.

a. Absorbed Dose on Thyroid in Japan for the first year Web site.  

b. Estimated doses to evacuees in Japan for the first year Web site.

c. Effective doses in Japan for the first year Web site.

4. Internal thyroid doses to Fukushima residents—estimation and issues remaining. Kim E, Kurihara O, Kunishima N, Momose T, Ishikawa T, Akashi M. J Radiat Res. 2016 Aug; 57(Suppl 1): i118–i126.