- Email: [email protected]
Histologically proven acute tubular necrosis in a series of 27 ICU patients
Accepted Manuscript Histologically proven acute tubular necrosis in a series of 27 ICU patients
René Robert, Denis Frasca, Bertrand Souweine, Jean-François Augusto, Carole Philipponnet, Florent Joly, Jean-Michel Goujon, Nicolas Lerolle PII: DOI: Reference:
S0883-9441(18)30801-3 doi:10.1016/j.jcrc.2018.08.029 YJCRC 53021
To appear in:
Journal of Critical Care
Please cite this article as: René Robert, Denis Frasca, Bertrand Souweine, Jean-François Augusto, Carole Philipponnet, Florent Joly, Jean-Michel Goujon, Nicolas Lerolle , Histologically proven acute tubular necrosis in a series of 27 ICU patients. Yjcrc (2018), doi:10.1016/j.jcrc.2018.08.029
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Histologically proven acute tubular necrosis in a series of 27 ICU patients
René Robert1, Denis Frasca 2, Bertrand Souweine3, Jean-François Augusto 4, Carole Philipponnet 3,
1 René
SC
RI
PT
Florent Joly1, Jean-Michel Goujon 5, Nicolas Lerolle4
Robert :
NU
Université de Poitiers, CHU Poitiers Médecine Intensive Réanimation, Poitiers, France [email protected]
MA
F Joly : Université de Poitiers, CHU Poitiers Médecine Intensive Réanimation, Poitiers, France e-mail [email protected]
Frasca, Université de Poitiers, CHU Poitiers, Département d’Anesthésie-Réanimation, France
3B
EP
TE
e-mail [email protected]
D
2D
Souweine : Université d’Auvergne CHU de Clermont-Ferrand Service de Réanimation Médicale, Hôpital
AC C
Gabriel Montpied, Clermont-Ferrand, France e-mail : [email protected] C Philipponet : Université d’Auvergne CHU de Clermont-Ferrand Service de Réanimation Médicale, Hôpital Gabriel Montpied, Clermont-Ferrand, France e-mail [email protected]
4 JF Augusto
: Université d’Angers, CHU Angers, Service de Nephrologie-Dialyse-Transplantation, Angers,
France. Email : [email protected]
ACCEPTED MANUSCRIPT N Lerolle : Département de Médecine Intensive Réanimation et Médecine hyperbare, CHU Angers, Université d’Angers, Angers, France e-mail: [email protected]
5 JM Goujon
: Université de Poitiers, CHU Poitiers Laboratoire d’anatomopathologie, Poitiers, France
RI
PT
e-mail [email protected]
Corresponding author:
SC
René Robert :
NU
Université de Poitiers, CHU Poitiers Médecine Intensive Réanimation, Poitiers, France [email protected]
AC C
EP
TE
D
MA
(+33)549444367
AC C
EP
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Abstract Purpose: Since renal biopsy is rarely performed for identifying acute tubular necrosis in ICU patients, there is little information on the real histopathological abnormalities observed in such situations. Materials and Methods : The clinical data of 27 patients with a confirmed diagnostic of acute tubular necrosis issued from two recent series gathering 125 patients who had renal biopsy during their ICU stay
PT
were reviewed. They were divided into sepsis (n=14) and non-sepsis (n=13) groups. Histopathologic lesions were reanalyzed and semi-quantitatively graded by a pathologist without knowledge of clinical
RI
characteristics of the patients.
SC
Results : SAPS2 and SOFA scores were identical in the two groups. Half of the patients had neither sepsis
NU
nor shock. The histopathological score was higher in the septic than in the non-septic group: 9 [IC; 9-11] vs 7 [IC 5.25-8.75]; p=0.01. There was no striking histopathological difference between septic and non-septic
MA
patients. However, the cytotoxic edema score was higher (3 [1; 3] vs 1 [0 ; 1]; p=0.006), and interstitial infiltration with polymorphonuclears was more frequent (p=0.02) in septic than in non-septic patients.
D
Conclusions: Septic and non-septic ICU patients with ATN had similar histopathologic features but lesions
TE
were more severe than in septic than in non-septic patients.
AC C
EP
Key words: acute tubular necrosis, renal biopsy, acute renal injury, histopathologic
List of abbreviations:
AKI: Acute kidney injury ARF: Acute renal failure ICU: Intensive care unit
ATN: Acute tubular injury RBF: Renal blood flow RB: Renal biopsy SAPS: Simplified acute physiologic score
ACCEPTED MANUSCRIPT SOFA: Sepsis-related organ failure assessment KDIGO: Kidney disease: improving global outcomes S: sepsis NS: non-sepsis
AC C
EP
TE
D
MA
NU
SC
RI
PT
SIRS: Systemic inflammatory response syndrome
ACCEPTED MANUSCRIPT
Introduction For years, tubular necrosis has been considered as the main mechanism for acute renal injury (AKI) in ICU patients [1,2]. Ruling out the rare possibilities of tubular obstruction occurring in specific conditions (i.e. rhabdomyolysis, drug crystal tubular precipitation, and myeloma), the tubular necrosis mechanism is
process [3].
And yet, interestingly, notwithstanding the
PT
usually mainly attributed on the basis of macrohemodynamic changes in the warm ischemia -reperfusion paucity of clinical studies providing
RI
histopathological information and based on experimental animal studies and human data obtained post-
SC
mortem, the classical mechanism of ischemia-reperfusion in patients with acute tubular necrosis (ATN) has
NU
been challenged, especially in patients with septic shock [4-6]. It has been shown that the renal lesions associated with AKI in septic shock are more complex than the classical ischemia-reperfusion mechanism.
MA
Thus, the occurrence of sepsis-induced AKI is not exclusive of decreased renal blood flow (RBF) states; in addition, it could develop in the setting of increased RBF [7]. Moreover, experimental models have varied experimental conditions [5]. Some
D
and induced different histopathologic alterations according to
TE
experimental models of septic shock with AKI have even reported an absence of renal histopathological alteration [8]. Several concerns and concepts have consequently been voiced and put forward [6,9]. In situations, renal injury cannot be explained on the sole basis of the classic paradigm of
EP
some
AC C
hypoperfusion. In fact, three main alterations may contribute to AKI during sepsis: inflammation, diffuse microcirculatory flow abnormalities, and cell bio-energetic adaptive responses to injury [6]. Most of the studies aimed at better understanding histopathological abnormalities during septic shock have been performed post-mortem [10-13]. They have shown patchy, heterogeneous tubular cell injury with apical vacuolization, but without tubular necrosis, foci of intra-luminal calcium phosphate crystals or extensive apoptosis [12,13]. In ICU, renal biopsy (RB) attempts to identify causes other than ATN, such as acute interstitial nephritis or glomerulonephritis or to identify the reason for non-recovery of function [3,14]. There are consequently and logically few studies with focused on histopathological examination ATN identification [14]. Recently, two retrospective multicenter French studies reported RB data involving 125 ICU patients [15,16]. Following that, we collected the data of a subgroup of patients from these two series
ACCEPTED MANUSCRIPT with a histopathological diagnosis of ATN with the objective of analyzing the different lesions observed. We aimed to bring some information for the ATN histologic pattern in ICU patients and particularly to focus on the possibility to evidence differences in ATN in septic and in non-septic patients.
PT
Material and Methods: The clinical data of 31 patients with a confirmed diagnostic of non-specific ATN issued from two recent
RI
published French series bringing together 125 patients who had RB during their ICU stay were reviewed
SC
[15,16]. In these series, RB was usually performed either to facilitate etiologic diagnosis of acute renal failure (ARF), or because there had been no clinical improvement of ARF [15,16]. Patients with ATN
NU
associated with a major diagnosed pathology (e.g. vascularitis, thrombotic microangiopathy) were
MA
excluded from the analysis Clinical characteristics of the included patients were age, sex, SAPS2 on admission, SOFA score, shock, catecholamine infusion, use of invasive mechanical ventilation and sedation,
D
antibiotic administration. Major comorbidities including diabetes, arterial hypertension, ischemic
TE
cardiopathy and risk factors such as nephrotoxic agents (non-steroid anti-inflamatory agent, contrast injectio n, angiotensin-converting-enzyme inhibitor, alpha-2 adrenergic receptor agonist, aminoglycoside,
EP
vancomycin, amphotericin B) administration were reported. AKI was defined within 24 hours after ICU
AC C
admission according to the KDIGO criteria [3]. All patients had no significant known previous renal insufficiency defined with a creatinin plasma level below 130 µmol/l. Patients were divided into two groups: sepsis (S) and non-sepsis (NS) by the two physicians in charge of the study (RR and NL), who were unaware of the RB findings.
Sepsis was defined as clinically and microbiologically proven, or when suspected infection was associated with organ dysfunction [17]. RB was arbitrarily considered to have been performed early when it took place within 10 days of ICU admission and to have been performed late once this delay had elapsed. All the RBs were re-analyzed by a pathologist not involved in the previous two studies and without knowledge of the patients’ clinical characteristics. Light microscopic examination confirmed the ATN diagnostic and the histopathologic tubular lesions were analyzed by the pathologist, using a modified semi-quantitative
ACCEPTED MANUSCRIPT grading adapted from the Banff scoring system [18,19]. In summary, 6 basic histopathological lesions (four for tubular and two for interstitial injury evaluation) were graded from 0: no lesion to 4: major lesions. The sum total of the 6 lesions represented the histopathologic ATN score, with minimum and maximum values at 0 and 24. Eventual glomerular and microvascular cell infiltration and tubular vacuolisation were also reported, as was histopathological evidence of chronic ischemic lesion. Search for
tubular apoptotic
PT
deposit assessed by cell shrinkage, nucleus condensing and fragmenting seen in necrotic proximal tubules surrounding a glomerulus was systematically performed in all the biopsies. Chronic ischaemic changes
RI
were determined by severe interstitial fibrosis and tubular atrophy associated with glomeruli alterations.
SC
Neither electronic microscopy nor immunostaining were performed.
NU
The ethics committees of the University hospitals of Angers, France and Clermont-Ferrand, France
MA
approved the extraction of data from the patient charts.
Statistical analysis:
D
Values are as medians with first and third quartiles for continuous variables. Categorical variables are
TE
represented with number of patients by category. Subgroup analyses were performed according to delay (early e.g. within 10 days after admission or late, subsequent to this delay) of RB and to the value of the
EP
histopathological score (higher or lower than the median value).
AC C
Analyses included Fisher’s exact test for categorical variables and Mann-Whitney U test for continuous variables. All analyses were performed with R software 3.3.2. (R Foundation for Statistical Computing, Vienna, Austria). All p-values were two-tailed and p-values <0.05 were considered significant.
Results: Among the initially selected 31 patients, four patients were excluded : two patients with only medullar sample on RB and two patients due to technical reasons impeding analysis on RB. The clinical characteristics of the 27 patients finally included are summarized in Table 1. None of the patients had previous history of cirrhosis, myeloma or severe heart failure. Out of the 27 patients, 14 were
ACCEPTED MANUSCRIPT considered as non-septic and 13 as septic . The reasons for admission in both groups are listed in Table 2. Patients from the septic group were more frequently males and more frequently had shock; however, SAPS2 and SOFA score did not statistically differ between the two groups. Three patients from the nonseptic and five from the septic group were receiving renal replacement therapy when RB was performed. All the patients from the septic and 3 from the non-septic group received antibiotics (p<0.01). In the septic
PT
group, 9 patients received aminoglucoside and 6 received vancomycin. In the non-septic group, 2 patients received aminoglucoside and 2 vancomycin. None of the patients from either group received amphotericin
RI
B. Considering aminiglucoside, 3 patients had 1, 3 patients had 2 and 3 patients more than 2 injections (for
SC
two patients information was not available). RB was performed 5.5 (1-40) and 4 (2-33) days after admission
NU
in the non-septic and septic groups respectively. Nineteen out of 27 patients had RB performed early after admission (11 in non-septic and 8 in septic patients), whereas 8 patients had RB performed late (more than
MA
20 days after admission); 5 in the septic and 3 in the non-septic patients. The histopathological patterns of RB and, particularly, semi-quantitative analysis of tubulo-interstitial
D
lesions are summarized in Table 3. The histopathological severity score was higher in the septic than in the
TE
non-septic group (9 [CI; 9-11] vs 7 [CI 5.25;-8.75]; p=0.01). Among the septic patients, the three without
EP
shock had histopathological scores of 9, 7 and 13 respectively. Chronic ischemia lesions was observed in 8/13 patients with sepsis and 4/14 patients in the non-septic group (p=0.67). In glomeruli, lymphocytic
AC C
infiltration was rare in septic and in non-septic patients as well. As shown in Figure 1, which is given as an example of histologic findings, fine tubular vacuolisations were evidenced in half of the patients (5 nonseptic and 8 septic patients) and were not associated with colloid fluid infusion. There was no tubular apoptotic deposit. Significant interstitial infiltration occurred in 8 and 7 patients from the septic and nonseptic groups respectively. The cells involved in interstitial infiltration were mainly lymphocytes, as illustrated in Figure 1. However, in septic patients associated moderate polymorphonuclear (PMN) infiltration was more frequent than in non-septic patients 9/13 vs 3/14; p=0.02. In only one patient with very severe septic shock, RB found numerous calcium oxalate crystals in the lumen of the tubules. No eosinophil infiltration or granulomata was identified in the RB.
ACCEPTED MANUSCRIPT Comparing RB performed early or late, there was no specific histopathological difference (Table 4). Chronic tubular ischemia was identified in 9/18 of early and in 5/8 of late RBs (NS). Comparing the patients with histopathological scores lower or higher than the median value, no clinical phenotype was found and, notably, clinical severity SAPS2 and SOFA scores were similar (Table 5).
PT
Discussion:
RI
The main findings of our study were the following: First, half of the patients with histopathologic findings of NTA had neither sepsis nor shock; second, histologic lesions were very similar in both groups but overall
SC
severity of renal injury assessed by histopathologic score was significantly higher in septic than in non-
NU
septic patients and third, two histologic patterns were more marked in septic than in non-septic patients: a higher cytotoxic edema score, and more frequent moderate interstitial infiltration with PMNs.
MA
Most of the time, RB in ICU patients is indicated to identify specific causes of AKI, especially glomerular or vascular diseases, to evaluate its prognosis and to manage eventual specific therapy [20-21]. RB can also
D
help to understand why AKI persists [14,20,21]. Since no specific therapy is usually attempted, few studies
TE
have focused on the histopathologic findings of ATN. In our series, most of the patients from the non-septic group had no shock and NTA was an unexpected
EP
diagnosis. In these patients, there was no intra-tubular precipitation and no evidence of osmotic
AC C
nephropathy or toxic medication use. Interstitial infiltration was identified in 56% of the biopsies. This is in accordance with the inflammatory response associated with renal ischemia/reperfusion injury due to endothelial activation and leukocyte entrapment [22]. Few previous studies have reported that AKI could occur without overt signs of hypoperfusion [23,24]. There may exist particular vulnerability to moderate hypoperfusion when autoregulation is impaired, and it might help to explain normotensive ischemic acute renal failure [23]. In septic patients, besides
classical tubular injury involvement e.g. tubular dilation, loss of the brush
barrier, cytotoxic edema, and interstitial edema, other lesions such as interstitial infiltration, infiltration of the glomerular capillaries with monocytes and PMNs, apoptosis, autophagia, apical tubular vacuolisation
ACCEPTED MANUSCRIPT and tubular calcium crystal deposits have been reported [12,13,25]. However, in these studies lesions have been observed in immediate post-mortem RB and it may be difficult to differentiate the lesions induced by sepsis itself from those due to the agonic phase [12,13]. In our series, moderate interstitial lymphocytic infiltration and apical tubular vacuolisation were identified in half of the patients and were not more frequent in septic than in non-septic patients, suggesting that these phenomena were related more to
PT
global tubular injury mechanisms than to sepsis or shock. Tubular calcium crystal deposits were observed in only one septic patient of our series whereas they were observed in 50% of septic patients in post-mortem
RI
biopsy [13]. The potential physiopathologic role of this phenomenon was not investigated.
SC
In our study, there histologic findings were similar between septic and non-septic patients. Particularly, in
NU
both groups interstitial infiltration occurred in more than 50% of patients. The only significant differences between septic and non-septic patients consisted in the severity of cytotoxic edema and more frequent
MA
moderate interstitial infiltration with PMNs. During sepsis, PMNs may infiltrate renal parenchyma and cause tissue injury either through release of cytotoxic substances or impairment of microcirculatory flow
D
[14].
TE
In some patients, RB is usually performed late; this is mainly due to non-recovery recovery of renal function [14]. In this situation, additional renal injuries occurring during ICU stay or pre-existing chronic
EP
ischemia alterations might explain absence of recovery. In our study, interestingly, there was no
AC C
histopathological difference in acute ATN between the patients with or without pre-existing ischemic renal alteration. Additionally, the frequency of pre-existing renal ischemic lesions was similar in the septic and non-septic groups and in patients having their RB early or late. There are several limitations to this study: First, this was a retrospective study involving non-consecutive patients for whom the main target of RB was not to confirm ATN or to discriminate sepsic and non-septic patients. But interestingly, this retrospective review showed that ATN could be an unsuspected diagnostic in patients without classical risk factors of ATN, particularly in non-septic patients. Secondly, since most septic patients (10/13) and conversely only 3/14 of the patients with non-septic shock had associated shock, it was not possible to adjust our findings and to differentiate lesions induced by sepsis or by shock. Thirdly, several patients received aminoglycosid or vancomycin that might have contributed to the renal
ACCEPTED MANUSCRIPT injury. However, 5 of these patients received no more than two aminoglycosid injections, so that the contribution of antibiotic toxicity to the observed ATN was probably limited. Finally, we did not perform any additional immunostaining or electronic examination of the RB, even though it might have provided more specific information on potential histopathological differences in patients with or without sepsis. The main strengths of this study were the relatively high number of patients in this series of critically ill patients
PT
with RB performed during their ICU hospitalisation showing ATN patterns with semi-quantitative grading of the different components of the cellular alterations. This contrasts with the previous studies, which were
RI
mainly performed post-mortem. In one study, RB was performed in 40 septic ICU patients, showing ATN in
SC
50% of the cases; unfortunately, there was no detailed information concerning the histopathological lesions
NU
[26].
In summary, while analysis of RB of patients with ATN showed that septic and non-septic ICU patients had lesions were more severe in septic than in non-septic patients. The
MA
similar histopathologic features,
observed lesions were frequently not limited to the tubular injury, but also involved interstitial infiltration
D
suggesting that the term “acute tubulo-interstitial injury” would be more appropriate than the classical
TE
“acute tubular necrosis”. Further studies including immunostaining or electronic microscopy would be o f
AC C
EP
interest to confirm these results and to better understand the pathophysiological mechanisms involved.
ACCEPTED MANUSCRIPT
Authors' contributions RR and NL: design of the study JMG: histopathological analysis
PT
AC: statistical analysis
RI
RR, NL and BS: draft of the article
SC
RR, BS, NL, LP, JFA, FJ, JMG: reviewing the article
NU
Acknowledgment:
MA
We thank Jeffrey Arsham for reviewing and editing the manuscript. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-
TE
D
profit sectors.
EP
Conflict of interest:
AC C
The authors declare no conflict of interest Other Contributors:
Azoulay E, Canet E: Service de Réanimation Médicale, Hôpital Saint Louis, Paris, France; Bollaert PE : Service de Réanimation Médicale CHU de Nancy, 54035 Nancy, France; du Cheyron D : Service de Réanimation medicale, CHU de Caen, 14033 Caen, France; Beduneau G : Service de Re´animation medicale, CHU de Rouen, 76031 Rouen, France; Dijoud F : Laboratoire d’Anatomie Pathologie, Hospices Civils de Lyon, Lyon ; Fillatre P : Service de Réanimation médicale et maladies infectieuses, CHU de Rennes, 35033 Rennes, France; Guérin C :Service de Réanimation Médicale, Hospices Civils de Lyon, Université de Lyon, Lyon, France; Guitton C, Lassalle V : Service de Réanimation médicochirurgicale, Centre Hospitalier du
ACCEPTED MANUSCRIPT Mans, 72000 Le Mans, France; Heng AE : Service de Réanimation Médicale, CHU de Clermont-Ferrand, F63003, Clermont, Clermont-Ferrand, France; Mariat C : Service de Néphrologie, CHU Saint Etienne, Saint Etienne, France; Meziani F, Schenck-Dhif M : Service de Réanimation médicale, Hopitaux Universitaires de Strasbourg, 67091 Strasbourg, France; Perrotin D : Service de Réanimation médicale, CHU de Tours, 37044
AC C
EP
TE
D
MA
NU
SC
RI
PT
Tours, France; Vinsonneau C : Service de Réanimation médicale, Béthune, France
ACCEPTED MANUSCRIPT
References [1] Liano F, Pascual J. Epidemiology of acute renal failure: A prospective, multicenter, community-based study. Kidney Int 1996; 50: 811—818
PT
[2] Amdur RL, Chawla LS, Amodeo S, Kimmel PL, Palant CE. Outcomes following diagnosis of acute renal failure in U.S. veterans: focus on acute tubular necrosis. Kidney Int. 2009; 76: 1089–1097
SC
guideline for acute kidney injury. Kidney Int Suppl 2012;2:1–138.
RI
[3] The Kidney Disease Improving Global Outcomes (KDIGO) Working Group. KDIGO clinical practice
[4] Rosen S, Heyman SN. Difficulties in understanding human “acute tubular necrosis”: Limited data and
NU
flawed animal models. Kidney Int. 2001; 60: 1220–1224
[5] Heyman SN, Rosenberger C, Rosen C Experimental ischemia–reperfusion: biases and myths—the
MA
proximal vs. distal hypoxic tubular injury debate revisited. Kidney Int 2010; 77, 9–16 [6] Gomez H, Ince C, De Backer D, Pickkers P, Payen D, Hotchkiss J, Kellum JA. A unified theory of sepsis-
TE
adaptation injury. Shock 2014; 41:3-11
D
induced acute kidney injury: inflammation, microcirculatory dysfunction, bioenergetics and the tubular cell
9:R363-374.
EP
[7] Langenberg C, Bellomo R, May C, Wan L, Egi M, Morgera S: Renal blood flow in sepsis. Crit Care. 2005;
AC C
[8] Langenberg C, Glenda Gobe G, Hood S, May CN, Bellomo R. Renal Histopathology During Experimental Septic Acute Kidney Injury and Recovery. Crit Care Med 2014;42:e58–e67. [9] Bellomo R, Kellum JA, Ronco C. Acute kidney injury. Lancet 2012; 380: 756–66. [10] Sato T, Kamiyama Y, Jones RT, Cowley RA, Trump BF: Ultrastructural study on kidney cell injury following various types of shock in 26 immediate autopsy patients. Adv Shock Res. 1978, 1:55-69. [11] Hotchkiss RS, Swanson PE, Freeman BD, Tinsley KW, Cobb JP, Matuschak GM, et al. Apoptotic cell death in patients with sepsis, shock, and multiple organ dysfunction. Crit Care Med 1999;27:1230–1251. [12] Lerolle N, Nochy D,Guérot E, Bruneval P, Fagon JY, Diehl JL. Histopathology of septic shock induced acute kidney injury: apoptosis and leukocytic infiltration. Intensive Care Med 2010;36:471-478.
ACCEPTED MANUSCRIPT [13] Takasu O, Gaut JP, Watanabe E, To K, Fagley ER, Sato B, et al.
Mechanisms of Cardiac and Renal
Dysfunction in Patients Dying of Sepsis. Am J Respir Crit Care Med 2013; 187: 509–517 [14] Abdulkader RCRM, Libório AB, Malheiros DMAC. Histological Features of Acute Tubular Necrosis in Native Kidneys and Long-Term Renal Function. Renal Failure 2008; 30:667–673 [15] Augusto JF, Lassalle V,Fillatre P, Perrotin D, Meziani F, Schenck-Dhif M, Bollaert PE, et al. Safety and
PT
diagnostic yield of renal biopsy in the intensive care unit. Intensive Care Med 2012; 38:1826–1833 [16] Philliponet C , Guérin C , Canet E, Robert R, Mariat C, Dijoud F, et al. Kidney biopsy in the critically ill
RI
patient, results of a multicentre retrospective case series. Minerva Anestesiol 2013;79:53-61
SC
[17] Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, et al he Third
NU
International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 2016 ;315:801-10. [18] Goujon JM, Hauet T, Menet E, Levillain P, Pabin P, Carretier M. Histopathological evaluation of
MA
proximal tubule cell injury in isolated perfused pig kidneys exposed to cold ischemia. J Surgical Res 199;82:228-233
D
[19] Solez K, Axelsen RA, Benediktsson H, Burdick JF, Cohen AH, Colvin RB et al International
TE
standardization of criteria for the histologic diagnosis of renal allograft rejection: The Banff working classification of kidney transplant pathology. Kidney Int 1993;. 44: 411-422
AC C
Nephrol 1984;4:27–31.
EP
[20] Mustonen J, Pasternack A, Helin H, Pystynen S, Tuominen T. Renal biopsy in acute renal failure. Am J
[21] Andreucci VE, Fuiano G, Stanziale P, Andreucci M. Role of renal biopsy in the diagnosis and prognosis of acute renal failure. Kidney Int Suppl 1998;66:S91–5. [22] Bonventre JV, Zuk A. Ischemic acute renal failure: an inflammatory desease ? Kidney Int 2004; 66: 480485 [23] Abuelo JG. Normotensive ischemic acute renal failure. N Engl J Med 2007;357:797-805. [24] Murugan R, Karajala-Subramanyam V, Lee M, et al.: Acute kidney injury in non-severe pneumonia is associated with an increased immune response and lower survival. Kidney Int 2010;77: 527-535,. [25] Langenberg C, Bagshaw SM, May CN, Bellomo R. The histopathology of septic acute kidney injury: a systematic review. Crit Care 2008; 12:R38
ACCEPTED MANUSCRIPT [26] Diaz de Leon M, Moreno SA, Gonzalez Diaz DJ, Briones GJ: Severe sepsis as a cause of acute renal
AC C
EP
TE
D
MA
NU
SC
RI
PT
failure. Nefrologia 2006; 26:439-444.
ACCEPTED MANUSCRIPT
Figure 1 Histopathological changes in a patient with acute tubular necrosis associated with sepsis
A. Light microscopy Magnification X200 : Moderate interstitial mononuclear cell infiltrates; B. Light microscopy Magnifcation x200 Tubular edema, fine Proximal tubule vacuolisations; C. Polarized Light
AC C
EP
TE
D
MA
NU
SC
RI
PT
Magnification X200 Obstructive changes with oxalate crystal deposits
ACCEPTED MANUSCRIPT Table 1 : Clinical characteristics of patients from the non-septic and septic groups. Values for continuous
Non sepsis n=14
Sepsis n=13
P
Age mean, SD
55 [36 ; 73]
66 [55 ; 72]
0.35
Sex male (n)
6
0.01
Sex female (n)
8
PT
data are median [interquartile; range]
12
RI
1
4
0.38
Hypertension
5
5
1
Ischemic cardiopathy
1
2
0.60
45 [34 ; 55]
50 [40 ; 58]
0.60
8 [7 ; 10]
12 [9 ; 14]
0.07
3
10
0.028
Invasive mechanical ventilation (n)
8
9
0.7
Sedation (n)
8
9
1
3
13
<0.001
NU
2
MA
Diabetes
SC
Comorbidities (n)
SAPS2 mean, SD
D
SOFA mean, SD
TE
Events preceding RB:
AC C
EP
Shock (n)
Antibiotic therapy (n)
Potential nephrotoxic agents 7 days before RB (n)
0.79
NSAID (n)
2
1
Contrast injection
2
3
ACEI or A2RA
3
3
Aminoglycosides
0
1
Vancomycin
0
1
ACCEPTED MANUSCRIPT KDIGO on admission
1
KDIGO 1 (n)
3
2
KDIGO 2 (n)
0
1
KDIGO 3 (n)
11
10
KDIGO the day of RB
1
2
KDIGO 3 (n)
11
PT
KDIGO 2 (n)
1 3
RI
1
9
SC
KDIGO 1 (n)
NU
SAPS: Simplified Acute Physiologic Score; SOFA: Sequential Organ Failure; KDIGO: Kidney Disease Improving Global Outcomes; RB: renal biopsy ; NSAID: nonsteroidal anti-inflammatory drug; ACEI : Angiotensin-converting-enzyme
AC C
EP
TE
D
MA
inhibitor: A2ARA: alpha-2 Adrenergic Receptor Agonists;
Table 2: Reason of ICU admission in non septic and septic patients Non sepsis
Sepsis
Acute respiratory failure (n=2)
Sepsis severe, septic shock (n=8)
Cardiogenic shock (n=2)
Pneumonia, ARDS (n=3)
Seizure, coma (n=3)
Malaria (n=1)
Keto-acidosis (n=1)
Endocarditis (n=1)
ACCEPTED MANUSCRIPT HELLP syndrome (n=1) Acute hepatitis (n=1) Polyradiculoneuritis (n=1)
PT
Liver transplantation (n=1) Gastroenteritis (n=1)
AC C
EP
TE
D
MA
NU
SC
RI
Major asthenia (n=1)
ACCEPTED MANUSCRIPT Table 3: Histopathologic features (optical light microscopy) in the non-septic and septic groups. Each was scored from 0 (no lesion) to 4 (major lesions); see text. Values for continuous data are median [interquartile
Sepsis
(n=14)
(n=13)
Interstitial infiltrate
1.5 [0.25 ; 2]
2 [1 ; 3]
Interstitial edema
0.5 [0 ; 1]
1 [1 ; 1]
0.24
Brush border loss
2 [1.25 ; 3]
2 [2 ; 3]
0.13
Cytotoxic edema
1 [0 ; 1]
3 [1 ; 3]
0.006
Endoluminal desquamation
1 [1 ; 1.75]
1 [0 ; 1]
0.37
Tubular ischemia
1 [0.25 ; 2]
2 [1 ; 3]
0.25
Total histopathologic severity score
7 [5.25 ; 8.75]
9 [9 ; 11]
0.01
RI
NU
MA
D TE EP AC C
p
PT
Non sepsis
SC
range].
0.25
ACCEPTED MANUSCRIPT Table 4: histopathological features according to early or late renal biopsy. Values for continuous data are median [interquartile range].
Late RB
(n=18)
(n=8)
Interstitial infiltrate
2 [0.5 ; 3]
1 [1 ; 2]
0.68
Interstitial edema
1 [0 ; 1]
1 [0.75 ; 1]
0.62
Brush border loss
2 [2 ; 3]
2 [1.75 ; 2.25]
0.84
Cytotoxic edema
1 [0.5 ; 1.5]
Endoluminal desquamation
1 [0.5 ; 1]
1 [0.75 ; 2]
0.58
Tubular ischemia
1 [0 ; 2]
2 [1.75 ; 3]
0.05
Total histopathologic severity score
8 [7 ; 9]
10.5 [8.25 ; 11]
0.20
RI
3 [1.5 ; 3]
SC
NU
MA D TE EP AC C
PT
Early RB
p
0.16
ACCEPTED MANUSCRIPT Table 5 Clinical characteristics according to low (≤ 8) and high (> 8) histopathologic score Score > 8
n=13
n=13
Age mean, SD
57 ± 19
61 ± 15
0.56
Sex male (n)
8
10
0.69
Sex female (n)
5
4
SAPS2 mean, SD
48.5 ± 21.4
SOFA mean, SD
9.1 ± 4.1
KDIGO 1 (n)
2
KDIGO 2 (n)
0 11
9.7 ± 4.1
0.70
3
1
RI 1 10
MA
Comorbidities (n)
0.60
44.9 ± 13.6
SC
NU
KDIGO 3 (n)
P
PT
Score ≤ 8
3
3
1
4
6
0.69
0
3
0.22
4
9
0.12
Invasive mechanical ventilation (n)
6
11
0.12
Sedation (n)
6
11
0.22
Antibiotic therapy (n)
4
9
0.13
Contrast injection 7 days before RB (n)
2
3
1
Diabetes
D
Hypertension
TE
Ischemic cardiopathy
EP
Events preceding RB
AC C
Shock (n)
SAPS: Simplified Acute Physiologic Score; SOFA: Sequential Organ Failure; KDIGO: Kidney Disease Improving
ACCEPTED MANUSCRIPT
Highlights
There is no specific histological feature to discriminate acute tubular necrosis in septic and non septic patients
Moderate interstitial infiltration is frequently associated with tubular injury both in septic and non
RI
Histopathologic alterations in acute tubular necrosis are more severe in septic than in non septic
EP
TE
D
MA
NU
SC
patients
AC C
PT
septic patients
Figure 1
René Robert, Denis Frasca, Bertrand Souweine, Jean-François Augusto, Carole Philipponnet, Florent Joly, Jean-Michel Goujon, Nicolas Lerolle PII: DOI: Reference:
S0883-9441(18)30801-3 doi:10.1016/j.jcrc.2018.08.029 YJCRC 53021
To appear in:
Journal of Critical Care
Please cite this article as: René Robert, Denis Frasca, Bertrand Souweine, Jean-François Augusto, Carole Philipponnet, Florent Joly, Jean-Michel Goujon, Nicolas Lerolle , Histologically proven acute tubular necrosis in a series of 27 ICU patients. Yjcrc (2018), doi:10.1016/j.jcrc.2018.08.029
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Histologically proven acute tubular necrosis in a series of 27 ICU patients
René Robert1, Denis Frasca 2, Bertrand Souweine3, Jean-François Augusto 4, Carole Philipponnet 3,
1 René
SC
RI
PT
Florent Joly1, Jean-Michel Goujon 5, Nicolas Lerolle4
Robert :
NU
Université de Poitiers, CHU Poitiers Médecine Intensive Réanimation, Poitiers, France [email protected]
MA
F Joly : Université de Poitiers, CHU Poitiers Médecine Intensive Réanimation, Poitiers, France e-mail [email protected]
Frasca, Université de Poitiers, CHU Poitiers, Département d’Anesthésie-Réanimation, France
3B
EP
TE
e-mail [email protected]
D
2D
Souweine : Université d’Auvergne CHU de Clermont-Ferrand Service de Réanimation Médicale, Hôpital
AC C
Gabriel Montpied, Clermont-Ferrand, France e-mail : [email protected] C Philipponet : Université d’Auvergne CHU de Clermont-Ferrand Service de Réanimation Médicale, Hôpital Gabriel Montpied, Clermont-Ferrand, France e-mail [email protected]
4 JF Augusto
: Université d’Angers, CHU Angers, Service de Nephrologie-Dialyse-Transplantation, Angers,
France. Email : [email protected]
ACCEPTED MANUSCRIPT N Lerolle : Département de Médecine Intensive Réanimation et Médecine hyperbare, CHU Angers, Université d’Angers, Angers, France e-mail: [email protected]
5 JM Goujon
: Université de Poitiers, CHU Poitiers Laboratoire d’anatomopathologie, Poitiers, France
RI
PT
e-mail [email protected]
Corresponding author:
SC
René Robert :
NU
Université de Poitiers, CHU Poitiers Médecine Intensive Réanimation, Poitiers, France [email protected]
AC C
EP
TE
D
MA
(+33)549444367
AC C
EP
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Abstract Purpose: Since renal biopsy is rarely performed for identifying acute tubular necrosis in ICU patients, there is little information on the real histopathological abnormalities observed in such situations. Materials and Methods : The clinical data of 27 patients with a confirmed diagnostic of acute tubular necrosis issued from two recent series gathering 125 patients who had renal biopsy during their ICU stay
PT
were reviewed. They were divided into sepsis (n=14) and non-sepsis (n=13) groups. Histopathologic lesions were reanalyzed and semi-quantitatively graded by a pathologist without knowledge of clinical
RI
characteristics of the patients.
SC
Results : SAPS2 and SOFA scores were identical in the two groups. Half of the patients had neither sepsis
NU
nor shock. The histopathological score was higher in the septic than in the non-septic group: 9 [IC; 9-11] vs 7 [IC 5.25-8.75]; p=0.01. There was no striking histopathological difference between septic and non-septic
MA
patients. However, the cytotoxic edema score was higher (3 [1; 3] vs 1 [0 ; 1]; p=0.006), and interstitial infiltration with polymorphonuclears was more frequent (p=0.02) in septic than in non-septic patients.
D
Conclusions: Septic and non-septic ICU patients with ATN had similar histopathologic features but lesions
TE
were more severe than in septic than in non-septic patients.
AC C
EP
Key words: acute tubular necrosis, renal biopsy, acute renal injury, histopathologic
List of abbreviations:
AKI: Acute kidney injury ARF: Acute renal failure ICU: Intensive care unit
ATN: Acute tubular injury RBF: Renal blood flow RB: Renal biopsy SAPS: Simplified acute physiologic score
ACCEPTED MANUSCRIPT SOFA: Sepsis-related organ failure assessment KDIGO: Kidney disease: improving global outcomes S: sepsis NS: non-sepsis
AC C
EP
TE
D
MA
NU
SC
RI
PT
SIRS: Systemic inflammatory response syndrome
ACCEPTED MANUSCRIPT
Introduction For years, tubular necrosis has been considered as the main mechanism for acute renal injury (AKI) in ICU patients [1,2]. Ruling out the rare possibilities of tubular obstruction occurring in specific conditions (i.e. rhabdomyolysis, drug crystal tubular precipitation, and myeloma), the tubular necrosis mechanism is
process [3].
And yet, interestingly, notwithstanding the
PT
usually mainly attributed on the basis of macrohemodynamic changes in the warm ischemia -reperfusion paucity of clinical studies providing
RI
histopathological information and based on experimental animal studies and human data obtained post-
SC
mortem, the classical mechanism of ischemia-reperfusion in patients with acute tubular necrosis (ATN) has
NU
been challenged, especially in patients with septic shock [4-6]. It has been shown that the renal lesions associated with AKI in septic shock are more complex than the classical ischemia-reperfusion mechanism.
MA
Thus, the occurrence of sepsis-induced AKI is not exclusive of decreased renal blood flow (RBF) states; in addition, it could develop in the setting of increased RBF [7]. Moreover, experimental models have varied experimental conditions [5]. Some
D
and induced different histopathologic alterations according to
TE
experimental models of septic shock with AKI have even reported an absence of renal histopathological alteration [8]. Several concerns and concepts have consequently been voiced and put forward [6,9]. In situations, renal injury cannot be explained on the sole basis of the classic paradigm of
EP
some
AC C
hypoperfusion. In fact, three main alterations may contribute to AKI during sepsis: inflammation, diffuse microcirculatory flow abnormalities, and cell bio-energetic adaptive responses to injury [6]. Most of the studies aimed at better understanding histopathological abnormalities during septic shock have been performed post-mortem [10-13]. They have shown patchy, heterogeneous tubular cell injury with apical vacuolization, but without tubular necrosis, foci of intra-luminal calcium phosphate crystals or extensive apoptosis [12,13]. In ICU, renal biopsy (RB) attempts to identify causes other than ATN, such as acute interstitial nephritis or glomerulonephritis or to identify the reason for non-recovery of function [3,14]. There are consequently and logically few studies with focused on histopathological examination ATN identification [14]. Recently, two retrospective multicenter French studies reported RB data involving 125 ICU patients [15,16]. Following that, we collected the data of a subgroup of patients from these two series
ACCEPTED MANUSCRIPT with a histopathological diagnosis of ATN with the objective of analyzing the different lesions observed. We aimed to bring some information for the ATN histologic pattern in ICU patients and particularly to focus on the possibility to evidence differences in ATN in septic and in non-septic patients.
PT
Material and Methods: The clinical data of 31 patients with a confirmed diagnostic of non-specific ATN issued from two recent
RI
published French series bringing together 125 patients who had RB during their ICU stay were reviewed
SC
[15,16]. In these series, RB was usually performed either to facilitate etiologic diagnosis of acute renal failure (ARF), or because there had been no clinical improvement of ARF [15,16]. Patients with ATN
NU
associated with a major diagnosed pathology (e.g. vascularitis, thrombotic microangiopathy) were
MA
excluded from the analysis Clinical characteristics of the included patients were age, sex, SAPS2 on admission, SOFA score, shock, catecholamine infusion, use of invasive mechanical ventilation and sedation,
D
antibiotic administration. Major comorbidities including diabetes, arterial hypertension, ischemic
TE
cardiopathy and risk factors such as nephrotoxic agents (non-steroid anti-inflamatory agent, contrast injectio n, angiotensin-converting-enzyme inhibitor, alpha-2 adrenergic receptor agonist, aminoglycoside,
EP
vancomycin, amphotericin B) administration were reported. AKI was defined within 24 hours after ICU
AC C
admission according to the KDIGO criteria [3]. All patients had no significant known previous renal insufficiency defined with a creatinin plasma level below 130 µmol/l. Patients were divided into two groups: sepsis (S) and non-sepsis (NS) by the two physicians in charge of the study (RR and NL), who were unaware of the RB findings.
Sepsis was defined as clinically and microbiologically proven, or when suspected infection was associated with organ dysfunction [17]. RB was arbitrarily considered to have been performed early when it took place within 10 days of ICU admission and to have been performed late once this delay had elapsed. All the RBs were re-analyzed by a pathologist not involved in the previous two studies and without knowledge of the patients’ clinical characteristics. Light microscopic examination confirmed the ATN diagnostic and the histopathologic tubular lesions were analyzed by the pathologist, using a modified semi-quantitative
ACCEPTED MANUSCRIPT grading adapted from the Banff scoring system [18,19]. In summary, 6 basic histopathological lesions (four for tubular and two for interstitial injury evaluation) were graded from 0: no lesion to 4: major lesions. The sum total of the 6 lesions represented the histopathologic ATN score, with minimum and maximum values at 0 and 24. Eventual glomerular and microvascular cell infiltration and tubular vacuolisation were also reported, as was histopathological evidence of chronic ischemic lesion. Search for
tubular apoptotic
PT
deposit assessed by cell shrinkage, nucleus condensing and fragmenting seen in necrotic proximal tubules surrounding a glomerulus was systematically performed in all the biopsies. Chronic ischaemic changes
RI
were determined by severe interstitial fibrosis and tubular atrophy associated with glomeruli alterations.
SC
Neither electronic microscopy nor immunostaining were performed.
NU
The ethics committees of the University hospitals of Angers, France and Clermont-Ferrand, France
MA
approved the extraction of data from the patient charts.
Statistical analysis:
D
Values are as medians with first and third quartiles for continuous variables. Categorical variables are
TE
represented with number of patients by category. Subgroup analyses were performed according to delay (early e.g. within 10 days after admission or late, subsequent to this delay) of RB and to the value of the
EP
histopathological score (higher or lower than the median value).
AC C
Analyses included Fisher’s exact test for categorical variables and Mann-Whitney U test for continuous variables. All analyses were performed with R software 3.3.2. (R Foundation for Statistical Computing, Vienna, Austria). All p-values were two-tailed and p-values <0.05 were considered significant.
Results: Among the initially selected 31 patients, four patients were excluded : two patients with only medullar sample on RB and two patients due to technical reasons impeding analysis on RB. The clinical characteristics of the 27 patients finally included are summarized in Table 1. None of the patients had previous history of cirrhosis, myeloma or severe heart failure. Out of the 27 patients, 14 were
ACCEPTED MANUSCRIPT considered as non-septic and 13 as septic . The reasons for admission in both groups are listed in Table 2. Patients from the septic group were more frequently males and more frequently had shock; however, SAPS2 and SOFA score did not statistically differ between the two groups. Three patients from the nonseptic and five from the septic group were receiving renal replacement therapy when RB was performed. All the patients from the septic and 3 from the non-septic group received antibiotics (p<0.01). In the septic
PT
group, 9 patients received aminoglucoside and 6 received vancomycin. In the non-septic group, 2 patients received aminoglucoside and 2 vancomycin. None of the patients from either group received amphotericin
RI
B. Considering aminiglucoside, 3 patients had 1, 3 patients had 2 and 3 patients more than 2 injections (for
SC
two patients information was not available). RB was performed 5.5 (1-40) and 4 (2-33) days after admission
NU
in the non-septic and septic groups respectively. Nineteen out of 27 patients had RB performed early after admission (11 in non-septic and 8 in septic patients), whereas 8 patients had RB performed late (more than
MA
20 days after admission); 5 in the septic and 3 in the non-septic patients. The histopathological patterns of RB and, particularly, semi-quantitative analysis of tubulo-interstitial
D
lesions are summarized in Table 3. The histopathological severity score was higher in the septic than in the
TE
non-septic group (9 [CI; 9-11] vs 7 [CI 5.25;-8.75]; p=0.01). Among the septic patients, the three without
EP
shock had histopathological scores of 9, 7 and 13 respectively. Chronic ischemia lesions was observed in 8/13 patients with sepsis and 4/14 patients in the non-septic group (p=0.67). In glomeruli, lymphocytic
AC C
infiltration was rare in septic and in non-septic patients as well. As shown in Figure 1, which is given as an example of histologic findings, fine tubular vacuolisations were evidenced in half of the patients (5 nonseptic and 8 septic patients) and were not associated with colloid fluid infusion. There was no tubular apoptotic deposit. Significant interstitial infiltration occurred in 8 and 7 patients from the septic and nonseptic groups respectively. The cells involved in interstitial infiltration were mainly lymphocytes, as illustrated in Figure 1. However, in septic patients associated moderate polymorphonuclear (PMN) infiltration was more frequent than in non-septic patients 9/13 vs 3/14; p=0.02. In only one patient with very severe septic shock, RB found numerous calcium oxalate crystals in the lumen of the tubules. No eosinophil infiltration or granulomata was identified in the RB.
ACCEPTED MANUSCRIPT Comparing RB performed early or late, there was no specific histopathological difference (Table 4). Chronic tubular ischemia was identified in 9/18 of early and in 5/8 of late RBs (NS). Comparing the patients with histopathological scores lower or higher than the median value, no clinical phenotype was found and, notably, clinical severity SAPS2 and SOFA scores were similar (Table 5).
PT
Discussion:
RI
The main findings of our study were the following: First, half of the patients with histopathologic findings of NTA had neither sepsis nor shock; second, histologic lesions were very similar in both groups but overall
SC
severity of renal injury assessed by histopathologic score was significantly higher in septic than in non-
NU
septic patients and third, two histologic patterns were more marked in septic than in non-septic patients: a higher cytotoxic edema score, and more frequent moderate interstitial infiltration with PMNs.
MA
Most of the time, RB in ICU patients is indicated to identify specific causes of AKI, especially glomerular or vascular diseases, to evaluate its prognosis and to manage eventual specific therapy [20-21]. RB can also
D
help to understand why AKI persists [14,20,21]. Since no specific therapy is usually attempted, few studies
TE
have focused on the histopathologic findings of ATN. In our series, most of the patients from the non-septic group had no shock and NTA was an unexpected
EP
diagnosis. In these patients, there was no intra-tubular precipitation and no evidence of osmotic
AC C
nephropathy or toxic medication use. Interstitial infiltration was identified in 56% of the biopsies. This is in accordance with the inflammatory response associated with renal ischemia/reperfusion injury due to endothelial activation and leukocyte entrapment [22]. Few previous studies have reported that AKI could occur without overt signs of hypoperfusion [23,24]. There may exist particular vulnerability to moderate hypoperfusion when autoregulation is impaired, and it might help to explain normotensive ischemic acute renal failure [23]. In septic patients, besides
classical tubular injury involvement e.g. tubular dilation, loss of the brush
barrier, cytotoxic edema, and interstitial edema, other lesions such as interstitial infiltration, infiltration of the glomerular capillaries with monocytes and PMNs, apoptosis, autophagia, apical tubular vacuolisation
ACCEPTED MANUSCRIPT and tubular calcium crystal deposits have been reported [12,13,25]. However, in these studies lesions have been observed in immediate post-mortem RB and it may be difficult to differentiate the lesions induced by sepsis itself from those due to the agonic phase [12,13]. In our series, moderate interstitial lymphocytic infiltration and apical tubular vacuolisation were identified in half of the patients and were not more frequent in septic than in non-septic patients, suggesting that these phenomena were related more to
PT
global tubular injury mechanisms than to sepsis or shock. Tubular calcium crystal deposits were observed in only one septic patient of our series whereas they were observed in 50% of septic patients in post-mortem
RI
biopsy [13]. The potential physiopathologic role of this phenomenon was not investigated.
SC
In our study, there histologic findings were similar between septic and non-septic patients. Particularly, in
NU
both groups interstitial infiltration occurred in more than 50% of patients. The only significant differences between septic and non-septic patients consisted in the severity of cytotoxic edema and more frequent
MA
moderate interstitial infiltration with PMNs. During sepsis, PMNs may infiltrate renal parenchyma and cause tissue injury either through release of cytotoxic substances or impairment of microcirculatory flow
D
[14].
TE
In some patients, RB is usually performed late; this is mainly due to non-recovery recovery of renal function [14]. In this situation, additional renal injuries occurring during ICU stay or pre-existing chronic
EP
ischemia alterations might explain absence of recovery. In our study, interestingly, there was no
AC C
histopathological difference in acute ATN between the patients with or without pre-existing ischemic renal alteration. Additionally, the frequency of pre-existing renal ischemic lesions was similar in the septic and non-septic groups and in patients having their RB early or late. There are several limitations to this study: First, this was a retrospective study involving non-consecutive patients for whom the main target of RB was not to confirm ATN or to discriminate sepsic and non-septic patients. But interestingly, this retrospective review showed that ATN could be an unsuspected diagnostic in patients without classical risk factors of ATN, particularly in non-septic patients. Secondly, since most septic patients (10/13) and conversely only 3/14 of the patients with non-septic shock had associated shock, it was not possible to adjust our findings and to differentiate lesions induced by sepsis or by shock. Thirdly, several patients received aminoglycosid or vancomycin that might have contributed to the renal
ACCEPTED MANUSCRIPT injury. However, 5 of these patients received no more than two aminoglycosid injections, so that the contribution of antibiotic toxicity to the observed ATN was probably limited. Finally, we did not perform any additional immunostaining or electronic examination of the RB, even though it might have provided more specific information on potential histopathological differences in patients with or without sepsis. The main strengths of this study were the relatively high number of patients in this series of critically ill patients
PT
with RB performed during their ICU hospitalisation showing ATN patterns with semi-quantitative grading of the different components of the cellular alterations. This contrasts with the previous studies, which were
RI
mainly performed post-mortem. In one study, RB was performed in 40 septic ICU patients, showing ATN in
SC
50% of the cases; unfortunately, there was no detailed information concerning the histopathological lesions
NU
[26].
In summary, while analysis of RB of patients with ATN showed that septic and non-septic ICU patients had lesions were more severe in septic than in non-septic patients. The
MA
similar histopathologic features,
observed lesions were frequently not limited to the tubular injury, but also involved interstitial infiltration
D
suggesting that the term “acute tubulo-interstitial injury” would be more appropriate than the classical
TE
“acute tubular necrosis”. Further studies including immunostaining or electronic microscopy would be o f
AC C
EP
interest to confirm these results and to better understand the pathophysiological mechanisms involved.
ACCEPTED MANUSCRIPT
Authors' contributions RR and NL: design of the study JMG: histopathological analysis
PT
AC: statistical analysis
RI
RR, NL and BS: draft of the article
SC
RR, BS, NL, LP, JFA, FJ, JMG: reviewing the article
NU
Acknowledgment:
MA
We thank Jeffrey Arsham for reviewing and editing the manuscript. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-
TE
D
profit sectors.
EP
Conflict of interest:
AC C
The authors declare no conflict of interest Other Contributors:
Azoulay E, Canet E: Service de Réanimation Médicale, Hôpital Saint Louis, Paris, France; Bollaert PE : Service de Réanimation Médicale CHU de Nancy, 54035 Nancy, France; du Cheyron D : Service de Réanimation medicale, CHU de Caen, 14033 Caen, France; Beduneau G : Service de Re´animation medicale, CHU de Rouen, 76031 Rouen, France; Dijoud F : Laboratoire d’Anatomie Pathologie, Hospices Civils de Lyon, Lyon ; Fillatre P : Service de Réanimation médicale et maladies infectieuses, CHU de Rennes, 35033 Rennes, France; Guérin C :Service de Réanimation Médicale, Hospices Civils de Lyon, Université de Lyon, Lyon, France; Guitton C, Lassalle V : Service de Réanimation médicochirurgicale, Centre Hospitalier du
ACCEPTED MANUSCRIPT Mans, 72000 Le Mans, France; Heng AE : Service de Réanimation Médicale, CHU de Clermont-Ferrand, F63003, Clermont, Clermont-Ferrand, France; Mariat C : Service de Néphrologie, CHU Saint Etienne, Saint Etienne, France; Meziani F, Schenck-Dhif M : Service de Réanimation médicale, Hopitaux Universitaires de Strasbourg, 67091 Strasbourg, France; Perrotin D : Service de Réanimation médicale, CHU de Tours, 37044
AC C
EP
TE
D
MA
NU
SC
RI
PT
Tours, France; Vinsonneau C : Service de Réanimation médicale, Béthune, France
ACCEPTED MANUSCRIPT
References [1] Liano F, Pascual J. Epidemiology of acute renal failure: A prospective, multicenter, community-based study. Kidney Int 1996; 50: 811—818
PT
[2] Amdur RL, Chawla LS, Amodeo S, Kimmel PL, Palant CE. Outcomes following diagnosis of acute renal failure in U.S. veterans: focus on acute tubular necrosis. Kidney Int. 2009; 76: 1089–1097
SC
guideline for acute kidney injury. Kidney Int Suppl 2012;2:1–138.
RI
[3] The Kidney Disease Improving Global Outcomes (KDIGO) Working Group. KDIGO clinical practice
[4] Rosen S, Heyman SN. Difficulties in understanding human “acute tubular necrosis”: Limited data and
NU
flawed animal models. Kidney Int. 2001; 60: 1220–1224
[5] Heyman SN, Rosenberger C, Rosen C Experimental ischemia–reperfusion: biases and myths—the
MA
proximal vs. distal hypoxic tubular injury debate revisited. Kidney Int 2010; 77, 9–16 [6] Gomez H, Ince C, De Backer D, Pickkers P, Payen D, Hotchkiss J, Kellum JA. A unified theory of sepsis-
TE
adaptation injury. Shock 2014; 41:3-11
D
induced acute kidney injury: inflammation, microcirculatory dysfunction, bioenergetics and the tubular cell
9:R363-374.
EP
[7] Langenberg C, Bellomo R, May C, Wan L, Egi M, Morgera S: Renal blood flow in sepsis. Crit Care. 2005;
AC C
[8] Langenberg C, Glenda Gobe G, Hood S, May CN, Bellomo R. Renal Histopathology During Experimental Septic Acute Kidney Injury and Recovery. Crit Care Med 2014;42:e58–e67. [9] Bellomo R, Kellum JA, Ronco C. Acute kidney injury. Lancet 2012; 380: 756–66. [10] Sato T, Kamiyama Y, Jones RT, Cowley RA, Trump BF: Ultrastructural study on kidney cell injury following various types of shock in 26 immediate autopsy patients. Adv Shock Res. 1978, 1:55-69. [11] Hotchkiss RS, Swanson PE, Freeman BD, Tinsley KW, Cobb JP, Matuschak GM, et al. Apoptotic cell death in patients with sepsis, shock, and multiple organ dysfunction. Crit Care Med 1999;27:1230–1251. [12] Lerolle N, Nochy D,Guérot E, Bruneval P, Fagon JY, Diehl JL. Histopathology of septic shock induced acute kidney injury: apoptosis and leukocytic infiltration. Intensive Care Med 2010;36:471-478.
ACCEPTED MANUSCRIPT [13] Takasu O, Gaut JP, Watanabe E, To K, Fagley ER, Sato B, et al.
Mechanisms of Cardiac and Renal
Dysfunction in Patients Dying of Sepsis. Am J Respir Crit Care Med 2013; 187: 509–517 [14] Abdulkader RCRM, Libório AB, Malheiros DMAC. Histological Features of Acute Tubular Necrosis in Native Kidneys and Long-Term Renal Function. Renal Failure 2008; 30:667–673 [15] Augusto JF, Lassalle V,Fillatre P, Perrotin D, Meziani F, Schenck-Dhif M, Bollaert PE, et al. Safety and
PT
diagnostic yield of renal biopsy in the intensive care unit. Intensive Care Med 2012; 38:1826–1833 [16] Philliponet C , Guérin C , Canet E, Robert R, Mariat C, Dijoud F, et al. Kidney biopsy in the critically ill
RI
patient, results of a multicentre retrospective case series. Minerva Anestesiol 2013;79:53-61
SC
[17] Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, et al he Third
NU
International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 2016 ;315:801-10. [18] Goujon JM, Hauet T, Menet E, Levillain P, Pabin P, Carretier M. Histopathological evaluation of
MA
proximal tubule cell injury in isolated perfused pig kidneys exposed to cold ischemia. J Surgical Res 199;82:228-233
D
[19] Solez K, Axelsen RA, Benediktsson H, Burdick JF, Cohen AH, Colvin RB et al International
TE
standardization of criteria for the histologic diagnosis of renal allograft rejection: The Banff working classification of kidney transplant pathology. Kidney Int 1993;. 44: 411-422
AC C
Nephrol 1984;4:27–31.
EP
[20] Mustonen J, Pasternack A, Helin H, Pystynen S, Tuominen T. Renal biopsy in acute renal failure. Am J
[21] Andreucci VE, Fuiano G, Stanziale P, Andreucci M. Role of renal biopsy in the diagnosis and prognosis of acute renal failure. Kidney Int Suppl 1998;66:S91–5. [22] Bonventre JV, Zuk A. Ischemic acute renal failure: an inflammatory desease ? Kidney Int 2004; 66: 480485 [23] Abuelo JG. Normotensive ischemic acute renal failure. N Engl J Med 2007;357:797-805. [24] Murugan R, Karajala-Subramanyam V, Lee M, et al.: Acute kidney injury in non-severe pneumonia is associated with an increased immune response and lower survival. Kidney Int 2010;77: 527-535,. [25] Langenberg C, Bagshaw SM, May CN, Bellomo R. The histopathology of septic acute kidney injury: a systematic review. Crit Care 2008; 12:R38
ACCEPTED MANUSCRIPT [26] Diaz de Leon M, Moreno SA, Gonzalez Diaz DJ, Briones GJ: Severe sepsis as a cause of acute renal
AC C
EP
TE
D
MA
NU
SC
RI
PT
failure. Nefrologia 2006; 26:439-444.
ACCEPTED MANUSCRIPT
Figure 1 Histopathological changes in a patient with acute tubular necrosis associated with sepsis
A. Light microscopy Magnification X200 : Moderate interstitial mononuclear cell infiltrates; B. Light microscopy Magnifcation x200 Tubular edema, fine Proximal tubule vacuolisations; C. Polarized Light
AC C
EP
TE
D
MA
NU
SC
RI
PT
Magnification X200 Obstructive changes with oxalate crystal deposits
ACCEPTED MANUSCRIPT Table 1 : Clinical characteristics of patients from the non-septic and septic groups. Values for continuous
Non sepsis n=14
Sepsis n=13
P
Age mean, SD
55 [36 ; 73]
66 [55 ; 72]
0.35
Sex male (n)
6
0.01
Sex female (n)
8
PT
data are median [interquartile; range]
12
RI
1
4
0.38
Hypertension
5
5
1
Ischemic cardiopathy
1
2
0.60
45 [34 ; 55]
50 [40 ; 58]
0.60
8 [7 ; 10]
12 [9 ; 14]
0.07
3
10
0.028
Invasive mechanical ventilation (n)
8
9
0.7
Sedation (n)
8
9
1
3
13
<0.001
NU
2
MA
Diabetes
SC
Comorbidities (n)
SAPS2 mean, SD
D
SOFA mean, SD
TE
Events preceding RB:
AC C
EP
Shock (n)
Antibiotic therapy (n)
Potential nephrotoxic agents 7 days before RB (n)
0.79
NSAID (n)
2
1
Contrast injection
2
3
ACEI or A2RA
3
3
Aminoglycosides
0
1
Vancomycin
0
1
ACCEPTED MANUSCRIPT KDIGO on admission
1
KDIGO 1 (n)
3
2
KDIGO 2 (n)
0
1
KDIGO 3 (n)
11
10
KDIGO the day of RB
1
2
KDIGO 3 (n)
11
PT
KDIGO 2 (n)
1 3
RI
1
9
SC
KDIGO 1 (n)
NU
SAPS: Simplified Acute Physiologic Score; SOFA: Sequential Organ Failure; KDIGO: Kidney Disease Improving Global Outcomes; RB: renal biopsy ; NSAID: nonsteroidal anti-inflammatory drug; ACEI : Angiotensin-converting-enzyme
AC C
EP
TE
D
MA
inhibitor: A2ARA: alpha-2 Adrenergic Receptor Agonists;
Table 2: Reason of ICU admission in non septic and septic patients Non sepsis
Sepsis
Acute respiratory failure (n=2)
Sepsis severe, septic shock (n=8)
Cardiogenic shock (n=2)
Pneumonia, ARDS (n=3)
Seizure, coma (n=3)
Malaria (n=1)
Keto-acidosis (n=1)
Endocarditis (n=1)
ACCEPTED MANUSCRIPT HELLP syndrome (n=1) Acute hepatitis (n=1) Polyradiculoneuritis (n=1)
PT
Liver transplantation (n=1) Gastroenteritis (n=1)
AC C
EP
TE
D
MA
NU
SC
RI
Major asthenia (n=1)
ACCEPTED MANUSCRIPT Table 3: Histopathologic features (optical light microscopy) in the non-septic and septic groups. Each was scored from 0 (no lesion) to 4 (major lesions); see text. Values for continuous data are median [interquartile
Sepsis
(n=14)
(n=13)
Interstitial infiltrate
1.5 [0.25 ; 2]
2 [1 ; 3]
Interstitial edema
0.5 [0 ; 1]
1 [1 ; 1]
0.24
Brush border loss
2 [1.25 ; 3]
2 [2 ; 3]
0.13
Cytotoxic edema
1 [0 ; 1]
3 [1 ; 3]
0.006
Endoluminal desquamation
1 [1 ; 1.75]
1 [0 ; 1]
0.37
Tubular ischemia
1 [0.25 ; 2]
2 [1 ; 3]
0.25
Total histopathologic severity score
7 [5.25 ; 8.75]
9 [9 ; 11]
0.01
RI
NU
MA
D TE EP AC C
p
PT
Non sepsis
SC
range].
0.25
ACCEPTED MANUSCRIPT Table 4: histopathological features according to early or late renal biopsy. Values for continuous data are median [interquartile range].
Late RB
(n=18)
(n=8)
Interstitial infiltrate
2 [0.5 ; 3]
1 [1 ; 2]
0.68
Interstitial edema
1 [0 ; 1]
1 [0.75 ; 1]
0.62
Brush border loss
2 [2 ; 3]
2 [1.75 ; 2.25]
0.84
Cytotoxic edema
1 [0.5 ; 1.5]
Endoluminal desquamation
1 [0.5 ; 1]
1 [0.75 ; 2]
0.58
Tubular ischemia
1 [0 ; 2]
2 [1.75 ; 3]
0.05
Total histopathologic severity score
8 [7 ; 9]
10.5 [8.25 ; 11]
0.20
RI
3 [1.5 ; 3]
SC
NU
MA D TE EP AC C
PT
Early RB
p
0.16
ACCEPTED MANUSCRIPT Table 5 Clinical characteristics according to low (≤ 8) and high (> 8) histopathologic score Score > 8
n=13
n=13
Age mean, SD
57 ± 19
61 ± 15
0.56
Sex male (n)
8
10
0.69
Sex female (n)
5
4
SAPS2 mean, SD
48.5 ± 21.4
SOFA mean, SD
9.1 ± 4.1
KDIGO 1 (n)
2
KDIGO 2 (n)
0 11
9.7 ± 4.1
0.70
3
1
RI 1 10
MA
Comorbidities (n)
0.60
44.9 ± 13.6
SC
NU
KDIGO 3 (n)
P
PT
Score ≤ 8
3
3
1
4
6
0.69
0
3
0.22
4
9
0.12
Invasive mechanical ventilation (n)
6
11
0.12
Sedation (n)
6
11
0.22
Antibiotic therapy (n)
4
9
0.13
Contrast injection 7 days before RB (n)
2
3
1
Diabetes
D
Hypertension
TE
Ischemic cardiopathy
EP
Events preceding RB
AC C
Shock (n)
SAPS: Simplified Acute Physiologic Score; SOFA: Sequential Organ Failure; KDIGO: Kidney Disease Improving
ACCEPTED MANUSCRIPT
Highlights
There is no specific histological feature to discriminate acute tubular necrosis in septic and non septic patients
Moderate interstitial infiltration is frequently associated with tubular injury both in septic and non
RI
Histopathologic alterations in acute tubular necrosis are more severe in septic than in non septic
EP
TE
D
MA
NU
SC
patients
AC C
PT
septic patients
Figure 1