Anti-TNFÎ± therapy early improves hemodynamics in

Journal of Crohn's and Colitis (2013) 7, 451–459

Available online at www.sciencedirect.com

Philippe Bonnin a, b,⁎, Jessica Coelho b, c , Marc Pocard b, d , Bernard I. Levy a, b , Philippe Marteau c a

Univ Paris Diderot, Sorbonne Paris Cité, AP-HP, Hôpital Lariboisière, Physiologie Clinique-Explorations-Fonctionnelles, 75010, Paris, France b Univ Paris Diderot, Sorbonne Paris Cité, INSERM, U965, 75010, Paris, France c Univ Paris Diderot, Sorbonne Paris Cité, AP-HP, Hôpital Lariboisière, hépato-gastro-entérologie, 75010, Paris, France d Univ Paris Diderot, Sorbonne Paris Cité, AP-HP, Hôpital Lariboisière, Chirurgie Digestive, 75010, Paris, France

Received 25 May 2012; received in revised form 3 July 2012; accepted 4 July 2012

KEYWORDS Crohn's disease; Anti-TNFα; Ultrasound imaging; Superior mesenteric artery; Retrobulbar arteries

Abstract Background and aims: Active Crohn's disease affects intestine but may alter other locations as eyes vasculature. Previous studies provide evidence of elevated blood flow velocities (BFv) and volume (BFV) in superior mesenteric artery (SMA). We prospectively studied hemodynamics in feeding arteries of bowel and eyes before and 2 weeks after treatment induction with anti-TNFα. Methods: Fifteen patients (5 females, 10 males, 35.4 ± 9.0 years, mean ± SD) with active Crohn's disease for 7.5 ± 7.7 years were enrolled. Ultrasound imaging was performed before and 2 weeks after treatment in SMA and retrobulbar arteries: central retinal (CRA), temporal posterior ciliary (TPCA) and ophthalmic (OA) arteries. Serum markers of inflammation (CRP and fibrinogen), arterial blood pressures (ABP) and skin flow-mediated dilation (sFMD) were measured and patients were compared to 10 control age- and sex-matched subjects. Results: Before treatment, CRP and fibrinogen plasma concentrations, SMA BFV (339 ± 100 mL/min) were higher in patients than in controls by 8.5-fold (p b 0.001), 1.4-fold (p b 0.01) and 1.5-fold, respectively (pb 0.01). BFv in CRA (3.5 ± 0.7 cm/s) and TPCA (4.4 ± 1.0 cm/s), sFMD (371 ± 469%) were significantly lower than in controls by 83%, 73% and 52% respectively (p b 0.05). Two weeks after treatment, CRP and fibrinogen decreased, SMA BFV was normalized (230 ± 39 L/min, p b 0.01), BFv in CRA, TPCA and OA increased respectively to 4.0 ± 1.1 (pb 0.05), 5.2 ± 1.4 (p b 0.001), 8.9 ± 3 cm/s (p b 0.05). ABP and sFMD remained unchanged.

⁎ Corresponding author at: Hôpital Lariboisière, 2 rue Ambroise Paré, 75010 Paris, France. Tel.: + 33 149958095; fax: + 33 149958871. E-mail address: [email protected] (P. Bonnin). 1873-9946/$ - see front matter © 2012 European Crohn's and Colitis Organisation. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.crohns.2012.07.002

Downloaded from https://academic.oup.com/ecco-jcc/article-abstract/7/6/451/452262 by Centre de Doc Medico Pharmaceutique user on 04 December 2018

Anti-TNFα therapy early improves hemodynamics in local intestinal and extraintestinalcirculations in active Crohn's disease

452

P. Bonnin et al. Conclusions: In active Crohn's disease, a first anti-TNFα administration rapidly normalized concomitantly plasma inflammatory markers and blood-flows in the mesenteric and retrobulbar arteries without affecting blood pressure and endothelial function. © 2012 European Crohn's and Colitis Organisation. Published by Elsevier B.V. All rights reserved.

Several arguments suggest a role of the blood vessels in the pathogenesis of Crohn's disease. 1 Vascular lesions consist in granulomatous vasculitis within the submucosal vascular plexus leading to microvascular thrombosis, submucosal ischemia and neovascularisation. 2–6 These morphological microvascular changes are associated with alterations of the mesenteric vascular blood supply and with impaired splanchnic hemodynamics. 2–4 Ischemia in the submucosa may result in the opening of small collateral shunts bypassing segments of microvascular occlusion and this may increase the blood flows in the splanchnic arteries. 1 The bowel wall ischemia may be favored by several cytokines including the vasoconstrictor and platelet-activating inducer endothelin-1, the granulomatous inflammation and procoagulant inductor and inflammatory cell chemo-attractor Tumor Necrosis Factor (TNFα). 7,8 Moreover, protective effects of nitric oxide (NO) such as vasodilation, microvascular permeability reduction, platelet and neutrophil adhesion inhibition are lost in Crohn's disease, because NO metabolism is derived towards free radical generation by stimulation of the inducible nitric oxide synthetase leading to endothelial dysfunction. 9 In recent years, ultrasound (US) Doppler imaging has been found to be increasingly useful for the investigation of splanchnic hemodynamics in Crohn's disease. 10–13 It is not invasive, accurate and reproducible, avoids exposure to radiation, and can be easily repeated. Ultrasound studies include either measurements of fasting or postprandial superior mesenteric artery (SMA) blood flow volume or evaluation of the bowel wall thickness and its vascular pattern. 14 SMA blood flow volume level has been found correlated closely with clinical activity assessed with the Crohn's disease activity index (CDAI) and with biological markers of inflammatory disease activity as C reactive protein. 10,15 Ultrasound imaging is easy to use with endoscopy and inflammatory markers to assess of the effects of treatments especially biological anti-TNFα therapy. 14 Monoclonal antibodies to Tumor Necrosis Factor (anti-TNFα) drugs provided major advances in the medical treatment of Crohn's disease as well as other inflammatory diseases sometimes associated with ocular inflammatory lesions. 16,17 Improvements of biological and clinical markers have been described in response to steroids and to thiopurines but never to anti-TNFα agents which have a rapid and intense antiinflammatory effect in Crohn's disease. Our goal was here to evaluate the early impact of a first anti-TNFα treatment in patients with Crohn's disease assessed by the evaluation of the hemodynamic changes in the superior mesenteric artery as well as in the clinical–biological responses. We also wondered whether endothelial function as well as hemodynamics in the retrobulbar arteries of the eye could be improved. We thus evaluated the flow-mediated vasodilation by noninvasive laser Doppler flowmetry (LDF) measurements in the skin microcirculation in response to a

local transdermal iontophoretic application of acetylcholine. We thus measured by ultrasound imaging the blood flow changes in the superior mesenteric artery, the ophthalmic, ciliary and central retinal arteries, in patients with active Crohn's disease, before and 2 weeks after induction of treatment with anti‐TNFα drugs.

2. Materials and methods 2.1. Patients and healthy controls Fifteen patients (5 females, 10 males) aged from 34.9± 9.1 [22–50] years (mean ±SD [min–max]) with active Crohn's disease from 5.1 ±4.9 [1–15] years were recruited for a prospective study consisting in systemic and local vascular supply hemodynamic parameter collections before and 2 weeks after a single monoclonal antibody anti-TNFα treatment (infliximab or adalimumab), i.e. just before the second administration. Disease activity was assessed using the Crohn's disease activity index (CDAI); active disease. 18 Disease was located at ileum in 6 patients, at colon in 5, at ileocolon in 2. Two patients presented anoperineal lesions only at the moment of the exploration but they both presented previous ileocolon location. Moreover, one patient presented a latter ileocolon location 2 years after the study (Table 1). Two patients had to be excluded from superior mesenteric artery (SMA) data collection because of unsatisfactory visualization of the artery after treatment but their measurements from retrobulbar arteries were conserved for data collection. Two patients presented associated spondylarthropathy, 5 were smokers. All patients were naïve from anti-TNFα treatment and associated treatments were oral corticoids for 8, budesonide for 3 and thiopurine for 3. A complete cardiovascular examination was performed with body mass index calculation and resting arterial blood pressure measurements (blood pressure monitor, Colin Corporation, Komaki-city, Japan). Fasting serum lipids, glucose, total plasmatic protein, fibrinogen, C reactive protein, hematocrit, and platelets count were determined using standard laboratory assays procedures. History of familial or individual gastroenteric and cardiovascular or ophthalmic disease and treatments were collected for each participant (Table 1). Patients were compared to 10 healthy controls age and sex-matched. Informed written consent was obtained from each patient prior to the examination, and the ethical guidelines of the 1975 declaration of Helsinki were followed throughout.

2.2. Ultrasound study 2.2.1. Superior mesenteric artery ultrasound imaging Ultrasound imaging was performed at rest in both patients and normal controls in 45° seat position after an overnight fast using an echograph ANTARES (Siemens, Erlangen, Germany) equipped with a linear transducer type CH4-1 (3.5 MHz). The


1. Introduction

Hemodynamic effects of anti-TNFα in active Crohn’s disease Table 1

Patient characteristics at entry. 10/5 35.4 ± 9.0 7.5 ± 7.7

2 13 19 ± 23 4.00 ± 1.07

[22–50] [1–28]

[4–86] [2.44–5.62]

6 5 2 4 2 (SPA)

5 22.8 ± 3.5

[18.82–29.84]

Values are expressed as mean ± SD.SD, standard deviation; [min– max].

superior mesenteric artery (SMA) was studied in its long axis in the sagittal plane. Color-coded Doppler was activated for digestive arteries recognition in front of the aorta. M-mode was activated for measurement of systolic and diastolic diameters of the inner wall of the SMA. Steer mode was used to ensure measurement of diameters strictly perpendicular to the longitudinal axis of the artery. Pulsed Doppler was activated for blood flow velocity waveform acquisition 2 to 3 cm after the SMA origin but proximal to any side branches with adjustment of the sample volume to the arterial section. A cut-off filter of 100 Hz was applied to exclude possible artifacts from vessel wall motion. Great care was taken for angle correction application to calculate true velocities. The transducer was arranged for each acquisition to obtain a maximum Doppler signal and to maintain an angle of insonation below 60°. Peak systolic [Vmax], end-diastolic [Vmin], timeaverage maximal [Vmax–mean] and spatial-average–timeaverage [Vmean–mean] velocities were calculated from the spectral analysis of the Doppler blood flow velocity waveforms. The time-average maximal velocity (maximum velocity per unit time [Vmax–mean]) is a computer-derived value calculated by integrating all the areas under the envelop of the waveform. The spatial-average–time-average mean velocity per unit time [Vmean–mean] is the computed-derived value calculated by integrating the area under instantaneous mean velocity. Indeed, blood flow velocity profile is parabolic in peripheral vessels, thus total displacement of blood is represented by the displacement of all the layers of blood on the arterial section, not only the highest one in the center of the arterial section but also the lower ones near the arterial wall. The spatialaverage mean velocity is representative of the mean blood displacement on all the arterial section per unit time; the spatial-average–time-averaged mean blood flow velocity is representative of the blood displacement over the wall

arterial section and over time. Blood volume is calculated with Vmean–mean following the formula: h i BV ¼ ðVmean–mean⋅60Þ⋅ π; ðD=2Þ2 where BV is the blood volume in mL/min, Vmean–mean is the spatial-averaged–time-averaged mean blood flow velocity in cm/s, and D the SMA diameter in cm. 19 Each measurement was taken in triplicate during quiet respiration and averaged. 2.2.2. Retrobulbar arteries ultrasound imaging Ultrasound imaging was performed on both eyes using an echograph ANTARES (Siemens, Erlangen, Germany) equipped with a linear transducer type VFX13.5 (13 MHz) as previously described. 20 Briefly, transducer was applied on closed eyelids without exerting pressure on the bulb. Two-dimensional ultrasound imaging was used to identify the optic nerve as a landmark. Then, color-Doppler mode was activated and retrobulbar vessels were identified, drawn and localized on the screen by their color-coded flow: central retinal artery (CRA), temporal posterior ciliary artery (TPCA) and ophthalmic artery (OA). A pulsed Doppler sample was then placed on the longitudinal axis of each vessel and pulsed Doppler spectrum recorded with simple repositioning of the pulsed Doppler sample from the CRA to the TPCA, then the OA. The angle between the Doppler beam and the long axis of each vessel was corrected to calculate actual flow velocities. Spatial-average– time-average mean blood flow velocity (Vmean–mean) was calculated in the 3 studied arteries. To minimize the variability of all US measurements, all investigations were performed by a single investigator (P.B.). The examiner was not blinded with respect to patient's history, but unaware of the actual clinical, endoscopic, and laboratory activity state. Intra-observer variability has been tested previously. Two series of paired measurements separated by 2 min interval performed by the same investigator were compared. The relative (positive or negative) differences (Di) within each pair of measures were calculated. The agreement between these two measurements was estimated by the mean and the standard deviation of the differences (Di). Repeatability of blood flow velocity measurements was investigated in the 20 controls through a calculation of the repeatability coefficient (RC) (British Standards Institution Precision of Test Method, i.e. according to RC 2 = ΣDi 2/N, where N is the sample). 21 This coefficient is the standard deviation of the estimated difference between two repeated measurements. The RC values for intra-observer repeatability were 1.5 cm/s for the spatial-average–time-average mean blood flow velocity in SMA, 1.0 cm/s in the OA, 0.4 cm/s in the CRA, 0.7 cm/s, and in the TPCA. 2.2.3. Acetylcholine‐iontophoresis and laser Doppler skin blood flow for determination of the flow-mediated dilation Endothelium-dependent vasodilation of the microcirculation of the forearm skin was evaluated by acethylcholine (Ach) iontophoresis in combination with laser Doppler velocimetry (Periflux 5000 system, Perimed, Järfällä, Sweden). The drug delivery chamber was loaded with 80 μL of acetylcholine (hydrochloride salt, Sigma, Steinheim, Germany) dissolved in distilled water at 20 g/L (0.0001 mol/L). Mean laser Doppler signal, expressed in an arbitrary perfusion unit, was measured under basal conditions (4 min), after iontophoresis of 3 Ach


Gender (M/F) Age (years) Duration of disease (years) Disease activity CDAI b 150 (n) 150 to 450 (n) CRP (mg/L) Fibrinogen (g/L) Disease extension Ileum (n) Colon (n) Ileocolon (n) Anoperineal lesions (n) Extra digestive locations (n) Crohn's disease and cardiovascular risk factors Smokers (n) BMI

453

454

P. Bonnin et al.

2.2.4. Statistics All results are expressed as mean ± SD [min, max]. Paired Student-t test was used to analyze drug effect on biological parameters and absolute values of time-average mean blood flow velocities. Statistics and linear regression analysis were done using MedCalc Software (Mariakerke, Belgium). Powers of the tests were verified and were superior to 85% in the retrobulbar arteries and equal to 100% in the superior mesenteric artery (http://www.anastasts.fr/).

after treatment whereas total platelets count (324 ± 95 [194–557] 10 9per L) total protein (70 ± 7 [61–85] g/L), fibrinogen (4.00 ± 1.07 [2.44–5.50] g/L) and CRP (19 ± 23 [4–86] mg/L) decreased 2 weeks after anti-TNFα treatment to respectively 270 ± 67 [182–401] (p b 0.01), 67 ± 4 [59–75] g/L (p b 0.05), 2.91 ± 1.15 [2.01–3.55] g/L (p b 0.001) and 7 ± 4 [4–15] mg/L (p b 0.01). CDAI similarly decreased from 214 ± 90 to 87 ± 61 (p b 0.001) with oral assessment by patients of well being (Table 2).

3.2. Patients did not exhibit endothelial function improvement after treatment Patients exhibited lower maximal vasodilator responses measured by skin laser Doppler flowmetry to the acetylcholine‐iontophoresis (Ach-) as well as after heat hyperemia as compared to controls (Table 2). Maximal Ach‐vasodilation and maximal heat hyperemia were respectively 371 ± 469 [5–1801]% and 628 ± 607 [− 15–2068]% in patients before treatment and 719 ± 387 [162–1667]% (p b 0.05), 1241 ± 451 [478–2454]% (p b 0.01) in controls. No improvement of skin flow mediated dilation was found after treatment.

3. Results 3.1. Patients exhibited inflammatory profiles before and amended after treatment At baseline, i.e. before anti-TNFα treatment, BMI was 22.8 ± 3.5 [18.82–29.84], resting systolic — 124 ± 10 [108–143], diastolic — 73 ± 11 [44–96], mean blood pressure 90 ± 11 [65–115] mm Hg, and resting heart rate was 61 ± 7 [48–71]. Arterial blood pressures and heart rate did not change after treatment. Hematocrit (40.1 ± 5.5 [27.9–48.4]%), hemoglobin (13.3 ± 1.9 [9.0–16.8] g/L), did not change

Table 2

3.3. Treatment normalized the blood flow volume in superior mesenteric artery Blood flow velocities and volume measured in the superior mesenteric artery (SMA) before treatment were higher in patients as compared to controls, spatial-average–timeaverage mean blood flow velocity (Vmean–mean) reached 22.7 ± 7.0 [12.7–39.5] cm/s and blood flow volume 339 ± 00 [191–572] mL/min, while 15.5 ± 4.7 [9.5–21.1] cm/s (p b 0.05) and 225 ± 58 [125–290] mL/min in controls (p b 0.01) (Table 3). A strong positive linear regression was found between SMA

Effects of a first anti-TNFα administration on systemic and cardiovascular parameters. Healthy controls

CDAI Hematology Hemoglobin (g/L) Haematocrit (%) Platelet count (10 9/L) Systemic inflammatory markers Total plasmatic protein (g/L) CRP (mg/L) Fibrinogen (g/L) Cardiovascular systemic parameters Systolic arterial blood pressure (mm Hg) Diastolic arterial blood pressure (mm Hg) Mean arterial blood pressure (mm Hg) Heart rate (bpm) Ach-skin laser Doppler vasodilation Maximal Ach-mediated vasodilation (%) Heat-hyperemia vasodilation (%)

NC

Patients with Crohn's disease Before

p-value vs controls

214 ± 90

Two weeks after 87 ± 61

p-value after vs before

p-value vs controls

p b 0.001

NS

13.9 ± 2.3 43.3 ± 6.8 224 ± 75

13.3 ± 1.9 40.1 ± 5.5 324 ± 95

NS NS NS

13.3 ± 1.6 40.5 ± 4.6 270 ± 67

NS NS 0.006

NS NS NS

72 ± 5 2±3 2.81 ± 0.42

70 ± 7 19 ± 23 4.00 ± 1.07

NS b 0.001 0.004

67 ± 4 7±4 2.91 ± 1.15

0.048 0.026 p b 0.001

NS NS NS

117 ± 10 67 ± 8 83 ± 8 65 ± 9

124 ± 10 73 ± 11 90 ± 11 61 ± 7

NS NS NS NS

122 ± 9 75 ± 13 91 ± 12 59 ± 10

NS NS NS NS

NS NS NS NS

719 ± 387 1241 ± 451

371 ± 469 628 ± 607

0.023 0.003

364 ± 317 630 ± 389

NS NS

0.003 p b 0.001

Values are expressed as mean ± SD; SD, standard deviation; NC, non convenient.


doses (3 × 2 min, endothelium-dependent response) and after heat hyperemia (5 min, endothelium-independent response). Ach was delivered with an anodal current (0.1 mA for 10 s) at 2-min intervals. Hyperemia was induced using a heat probe that heated the skin to 44 °C for 5 min. The changes in laser Doppler signal after Ach and heating were expressed as percent values of the mean basal signal. We previously evidenced a close correlation between endothelium function assessed by the flow-dependent brachial artery dilatation (method of reference) and the laser Doppler response to acetylcholine. 22

Hemodynamic effects of anti-TNFα in active Crohn’s disease

455

blood flow volume and CDAI before treatment (R = 0.77, p = 0.002, Fig. 1). After treatment, blood flow velocities and blood flow volume decreased in patients to respectively 17.6±5.0 [11.7–27.3] cm/s (pb 0.001) and to 230±39 [179–314] mL/min (pb 0.01), without any changes in the inner arterial diameter, not different from values measured in controls (Table 3). A positive linear regression was also found between SMA blood flow volume and CDAI after treatment (R= 0.71, p = 0.007, Fig. 1). From the two patients who had only anoperineal lesions at the moment of the study, one decreased his blood flow volume from − 27% (his ileal CD was in remission at the time of anti-TNF according to MRI) and one did not present any modification.

the retrobulbar arteries of eyes, particularly in central retinal and ciliary arteries.

Spatial-average–time-average mean blood flow velocities (Vmean–mean) measured in the temporal posterior ciliary artery (TPCA) and the central retinal artery (CRA) were lower before treatment compared to healthy subjects. Ophthalmic artery (OA) did not exhibit such a difference. Velocities were increased after treatment: from 7.8 ± 2.1 [4.2–11.7] to 8.9 ± 3.1 [4.3–16.9] cm/s in OA (p b 0.05), from 4.4 ± 1.0 [2.8–6.8] to 5.2 ± 1.4 [3.1–9.3] cm/s in TPCA (p b 0.001) and from 3.5 ± 0.7 [2.1–5.0] to 4.0 ± 1.1 [2.3– 7.4] cm/s in CRA (p b 0.05), reaching values of healthy subjects. A negative linear regression was found between Vmean–mean in CRA and CDAI only before treatment (R = 0.55, p = 0.004) (Table 4, Fig. 2).

4. Discussion In patients with active Crohn's disease, before a first injection of monoclonal antibody anti‐TNFα therapy, blood flow velocities and volume in the superior mesenteric artery (SMA) were increased, blood flow velocities in retrobulbar arteries were decreased compared to control subjects. We evidenced here a correlation between the normalization of hemodynamics in SMA and the amendment of the inflammatory status 2 weeks after anti-TNFα therapy. Moreover, we highlight the concomitant normalization of hemodynamics in

Table 3

A high flow rate in the SMA has been repeatedly observed in active Crohn's disease as in the present study.10,15 Noticeable, in our series, among the 5 patients with colonic CD and without ileal lesions, 4 had these alterations in the SMA blood supply suggesting a common vascular reactivity despite different locations of the disease. A theory has also been proposed to try to conciliate this with the observation of a progressive microvascular ischemia. Several independent research groups have described the microvascular anatomy and the bowel wall microvascular pathology in the areas of the intestine affected by the disease. 23–25 Studies have consistently demonstrated a reduced microvascular volume during the acute phase progressing to obliterative arteritis with disease chronicity. 26 Micro angiographic studies have consistently demonstrated a reduced number of feeding vessels. Furthermore, those vessels often present architectural abnormalities, being tortuous and have varying caliber. Noticeably, these anatomical changes occur only in segments of involved bowel. 27 Flow is preferentially reduced in the submucosa and the muscularis propria. Mucosal blood flow is relatively preserved. 28 Blood flow volume in SMA could thus be increased consecutively to the inflammatory hyperemia in the whole bowel; concomitantly, occurrence of segmental parietal microvascular ischemia may produce focal ischemic lesions, angiogenesis, and opening of small collateral shunts. Nitric oxide (NO) produced by endothelial cells may be involved. It has significant protective effects on mucosal integrity, reducing microvascular permeability, relaxing gastrointestinal smooth muscle and inhibiting platelet and neutrophil adhesion. 29 However, in the state of L-arginine deficiency, secondary to an excess in inducible NO synthase activity, as documented in Crohn's disease, 30 NO metabolism results in peroxynitrate production. Ischemic conditions also result in free radical production from the stable end-products of nitric oxide metabolism, nitrate and nitrite. 31 Nevertheless, while flow‐mediated endothelial vasodilation was altered in our patients, we did not observe any improvement after anti-TNFα

Effects of a first anti-TNF a administration on blood flow velocities in superior mesenteric artery.

Vmax (cm/s) Vmin (cm/s) Vmax–mean (cm/s) Vmean–mean (cm/s) PI RI Intima–intima internal diameter (mm) Blood flow volume (mL/min)

Healthy controls

Patients with Crohn's disease Before

p-value vs controls

Two weeks after

p-value after vs before

p-value vs controls

110.4 ± 24.5 13.8 ± 4.1 26.6 ± 8.3 15.5 ± 4.7 3.82 ± 1.19 0.87 ± 0.03 5.6 ± 0.4 225 ± 58

141.9 ± 42.8 19.9 ± 10.4 37.7 ± 12.9 22.7 ± 7.0 3.38 ± 1.03 0.86 ± 0.06 5.7 ± 0.9 339 ± 100

0.048 NS 0.027 0.048 NS NS NS 0.005

119.5 ± 35.5 15.6 ± 5.2 29.5 ± 8.2 17.6 ± 5.0 3.67 ± 1.03 0.87 ± 0.05 5.4 ± 0.7 230 ± 39

0.033 0.048 0.001 b 0.001 NS NS NS 0.001

NS NS NS NS NS NS NS NS

Values are expressed as mean ± SD; SD, standard deviation; Vmax, peak systolic velocity; Vmin, end-diastolic velocity; Vmax–mean, time-average maximal velocity (maximum velocity per unit time); and Vmean–mean, spatial-average–time-average mean velocity per unit time. PI, pulsatility index and RI, resistance index.


3.4. Treatment normalized the blood flow volume in retrobulbar arteries

4.1. Microvascular impairment in Crohn's disease

456

P. Bonnin et al.

4.2. TNFα is involved in the pathophysiology of Crohn's disease TNFα has many actions in Crohn's disease which could contribute to the formation of occlusive microvascular pathology even in extra intestinal territories. 33,34 These include induction of ICAM-1 allowing the development of granulomatous inflammation, induction of the release of the endothelial procoagulant platelet activating factor (PAF), increased microvascular permeability and inflammatory cell chemo-attraction. 8,33–35 Raised serum fibrinogen levels and thrombophilia are often present in Crohn's disease. 36

4.3. Vascular and hemorheologic effects of anti-TNFα drugs TNFα production can be controlled by reducing its production, by neutralizing circulating TNFα or by blocking the TNFα receptor. Receptor blockade was achieved with anti‐ TNFα treatments. Pharmacologic and immune blockage of TNFα inhibits its vascular and tissue effects, thus can help to fight against the local microvascular thrombosis and again its consequence, the physiological shunts bypassing of the microvascular occlusions. The therapeutic role of TNFα antagonists also supports a primary role for TNFα. Anti-TNFα has been a major progress for the treatment of Crohn's disease. 37 Animal studies have shown that anti‐TNFα reduced colonic inflammation in experimentally induced colitis. 38 Chimeric mouse– human IgG TNFα antibodies have been shown in randomized placebo-controlled clinical trials to induce a clinical response greater than 60% in Crohn's disease patients. 16 Figure 1 Normalization of the mean blood flow velocity (A) and blood flow volume (B) in the superior mesenteric artery (SMA) 2 weeks after a first anti-TNFα administration (***p b 0.001). The anti-TNFα administration decreased spatial-average–timeaverage mean blood flow velocity (Vmean–mean) as well as blood flow volume towards values presented by the control subjects. C: before (W0) as well as 2 weeks after administration (W2), there was a positive linear regression between blood flow volumes measured in SMA and the CDAI (W0: R = 0.77, p = 0.002; W2: R = 0.73, p = 0.007). The higher the CDAI, more increased was the blood flow volume in SMA.

treatment which suggests the absence of impact of the endothelial function (dysfunction) on the blood flow velocities and volume changes. Endothelin-1 (ET-1) has been involved in the primary vasoconstriction attested by elevation in plasma concentration. Furthermore, endothelial ET-1 receptor expression is greatest in the submucosal plexus. It is possible that a primary increase in ET-1 endothelial receptor number or reactivity triggers the

4.4. Changes in the superior mesenteric artery territory Anti‐TNFα administration was followed by a strong and early decrease in blood flow velocities and volume in the superior mesenteric artery in our patients with active Crohn's disease. This early effect must be related with the anti-inflammatory effects on the bowel wall consecutive to the inhibition of the inflammatory cascade which conducts 1) to the amendment of the inflammatory hyperemia and 2) to the restoration of the microvascular perfusion of the submucosa through the inhibition of the procoagulant and thrombotic effects of anti‐TNFα drug leading to the collapse of the physiological shunts bypassing of the microvascular occlusions. Together these effects could lead to normalization of the intestinal vascular network and to the restoration of a normal blood flow supply, proportionally to the decrease in inflammation markers in accordance with previous studies. 39,40 This is keeping with recent data showing a decrease in microvascular density in the injured bowel after anti-TNFα therapy using contrast enhanced power Doppler. 14 We chose to study the effects after


ischemic process. Interestingly, a recent study on the correlation between ET-1 and vascular endothelial growth factor (VEGF) concluded that the reciprocal reaction of these factors, both significantly raised in inflammatory bowel disease, resulted in primary vasoconstriction with secondary tissue hypoxia and tertiary angiogenesis. 32

Hemodynamic effects of anti-TNFα in active Crohn’s disease Table 4

457

Effects of a first anti-TNFα administration on blood flow velocities in the retrobulbar arteries. Healthy controls

Before

p-value vs Two weeks p-value after vs p-value vs controls after before controls

12.8 ± 3.6 11.9 ± 2.4 4.6 ± 1.6 3.7 ± 0.8 7.3 ± 2.3 6.2 ± 1.3 4.2 ± 1.2 3.5 ± 0.6 1.16 ± 0.22 1.31 ± 0.25 0.64 ± 0.06 0.68 ± 0.05

NS 0.018 0.041 0.011 0.027 0.025

13.3 ± 3.8 4.4 ± 1.5 7.2 ± 2.2 4.1 ± 1.1 1.29 ± 0.32 0.67 ± 0.08

NS 0.035 0.023 0.007 NS NS

NS NS NS NS NS NS

17.2 ± 6.5 14.2 ± 3.2 5.8 ± 1.5 5.1 ± 1.6 9.7 ± 2.6 8.0 ± 2.2 6.0 ± 2.0 4.4 ± 1.0 1.17 ± 0.22 1.17 ± 0.33 0.66 ± 0.09 0.64 ± 0.07

0.021 NS 0.028 0.001 NS NS

16.5 ± 4.8 6.1 ± 1.8 9.6 ± 2.9 5.2 ± 1.4 1.09 ± 0.22 0.63 ± 0.07

0.011 0.002 b 0.001 b 0.001 NS NS

NS NS NS NS NS NS

34.2 ± 6.6 8.1 ± 2.5 15.6 ± 3.4 9.0 ± 1.8 1.71 ± 0.43 0.76 ± 0.07

NS NS NS NS NS NS

34.6 ± 12.6 8.9 ± 3.3 15.6 ± 5.3 9.0 ± 3.0 1.71 ± 0.65 0.74 ± 0.08

NS 0.008 NS 0.009 0.029 0.027

NS NS NS NS NS NS

32.6 ± 8.4 7.2 ± 3.0 13.9 ± 3.9 7.7 ± 2.1 1.91 ± 0.67 0.77 ± 0.09

Values are expressed as mean ± SD; SD, standard deviation; Vmax, peak systolic velocity; Vmin, end-diastolic velocity; Vmax-mean, time-average maximal velocity (maximum velocity per unit time); and Vmean–mean, spatial-average–time-average mean velocity per unit time. PI, pulsatility index and RI, resistance index.

Figure 2 Normalization of the spatial-average–time-average mean blood flow velocity (Vmean–mean) in the central retinal (A) in the temporal ciliary (B) and in the ophthalmic arteries (C) 2 weeks after a first anti-TNFα administration (*p b 0.05, ***p b 0.001). D: before administration (W0), there was a negative linear regression between Vmean–mean measured in the central retinal artery and CDAI (R = 0.55, p = 0.004). The higher the CDAI, more decreased was the Vmean–mean in the central retinal artery.


Doppler study in the central retinal artery Vmax (cm/s) Vmin (cm/s) Vmax–mean (cm/s) Vmean–mean (cm/s) PI RI Doppler study in the long posterior ciliary artery Vmax (cm/s) Vmin (cm/s) Vmax–mean (cm/s) Vmean–mean (cm/s) PI RI Doppler study in the ophthalmic artery Vmax (cm/s) Vmin (cm/s) Vmax–mean (cm/s) Vmean–mean (cm/s) PI RI

Patients with Crohn's disease

458 just 2 weeks of anti-TNFα in order to assess very early events and cannot speculate of the further improvement which may occur somewhat later (for example after the end of the induction phase of the treatment or after 6 months).

During inflammation, the decrease in perfusion of vascular networks could be related to an increase of plasma proteins with rheological relevance. Indeed, hyperfibrinogenemia increases plasma viscosity, erythrocyte aggregation, thus increases the vascular resistance in the vascular networks. 41 The monoclonal antibody anti-TNFα infliximab inhibits the pleiotropic actions of TNFα and are widely used for the treatment of rheumatoid arthritis, juvenile idiopathic arthritis, spondyloarthropathies, Crohn's disease, sarcoidosis and psoriasis with an acceptable safety profile. Up to two-thirds of patients develop ocular inflammation eventually leading to significant visual impairment. A pathogenic role of TNFα in ocular inflammatory conditions has recently emerged from small trials reporting preliminary results on the efficacy of these agents in patients with noninfectious uveitis, regardless of the origin of the disease. TNFα antagonists are useful in the treatment of ocular inflammation associated with all cited inflammatory diseases and were also beneficial in small numbers of patients with idiopathic uveitis or scleritis. The currently available data are nonrandomized and thus preliminary. However, the preliminary evidence points to a growing optimism for targeting TNFα in patients with ocular inflammation. 17,42,43 Anti-TNFα (infliximab) has been intravitreously injected in a model of laser-induced choroidal neovascularization. Infliximab reduces angiogenesis at low doses. 44 Moreover, anti-TNFα treatment reduced the increases in leukocyte rolling, adhesion, and vascular leakage in a rat model of inflammatory uveitis. These results suggest the involvement of TNFα in inflammatory uveitis and its potential use as a therapeutic agent in the reduction of ocular inflammation. 45 Our patients, without uveitis, without neovessels in their eyes, exhibited low blood flow velocities in all retrobulbar arteries before treatment, possibly consecutive to blood circulating TNFα. Effects of anti-TNFα restored all retrobulbar blood flow velocities in the normal ranges and could be the result of the inhibition of the first steps of the inflammatory process. Indeed, TNFα could involve an increase in microvascular resistance mediated by both microvascular constriction and raised blood viscosity consecutive to the well-known increase in platelet aggregation and in fibrinogen. 46 Anti-TNFα thus could lead to a decrease in the local microvascular resistance and to the restoration of normal blood flows in parallel to the CDAI reduction. Blood flow normalization was strongly significant in the ciliary arteries, which supply the chorioretina, composed of strongly anastomosed small vessels, the pigmentary epithelium and the sclera, and in the central retinal artery which supplies the internal layer of the retina.

5. Conclusions In patients with active Crohn's disease, we evidenced a correlation between the normalization of hemodynamics in the superior mesenteric artery and the amendment of the

inflammatory status 2 weeks after a first injection of monoclonal antibody anti-TNFα therapy. This finding could be related to the amendment of the inflammatory hyperemia in the bowel and to the normalization of the intestinal vascular network. Moreover, we highlight the concomitant normalization of hemodynamics in all the retrobulbar arteries of eyes. This finding was not related here with an improvement in flow-mediated dilation but could proceed from the normalization in hemorheologic properties of blood leading to the normalization of hemodynamic resistance and blood flows in the extra intestinal vascular networks, particularly in the eyes.

Conflict of interest Philippe Marteau has received consultancy honoraria and invitation to give lectures from the companies: Abbott, MSD, Schering Plough, and UCB. Other authors do not have conflicts of interest.

Statement of authorship PB participated in the design of the experiments, collection of data, analysis of data, and writing of the manuscript. JC participated in the design of the experiments, collection of data, analysis of data, and writing of the manuscript. MP participated in the design of the experiments, analysis of data, and writing of the manuscript. BL participated in the design of the experiments, analysis of data, and writing of the manuscript. PhM participated in the design of the experiments, analysis of data, and writing of the manuscript.

Acknowledgments Authors thank Michèle Pauty, Valérie Paplomatas and Agnès Dutrey for their helpful expert technical assistance.

References 1. Thornton M, Solomon MJ. Crohn's disease: in defense of a microvascular aetiology. Int J Colorectal Dis 2002;17:287–97. 2. Carr ND, Pullan BR, Schofield PF. Microvascular studies in nonspecific inflammatory bowel disease. Gut 1986;27:542–9. 3. Brahme F, Lindström C. A comparative radiographic and pathological study of intestinal vasoarchitecture in Crohn's disease and ulcerative colitis. Gut 1970;11:928–40. 4. Erikson U, Fagerberg S, Krause U, Olding L. Angiographic studies in Crohn's disease and in ulcerative colitis. AJR Am J Roentgenol 1970;110:385–92. 5. Bousen E, Reuter SR. Mesenteric angiography in the evaluation of inflammatory and neoplastic disease of intestine. Radiology 1966;87:1028–36. 6. Wakefield AJ, Sawyerr AM, Dhillon AP, Pittilo RM, Rowles PM, Lewis AA, et al. Pathogenesis of Crohn's disease: multifocal gastrointestinal infarction. Lancet 1989;2:1057–62. 7. Kanazawa S, Tsunoda T, Onuma E, Majima T, Kagiyama M, Kikuchi K. VEGF, basic-FGF, and TGF-beta in Crohn's disease and ulcerative colitis: a novel mechanism of chronic intestinal inflammation. Am J Gastroenterol 2001;96:822–8. 8. Braegger CP, Nicholls S, Murch SH, Stephens S, MacDonald TT. Tumour necrosis factor alpha in stool as a marker of intestinal inflammation. Lancet 1992;339:89–91.


4.5. Changes in the retrobulbar arteries

P. Bonnin et al.

459

9. Schreiber S, Nikolaus S, Hampe J. Activation of nuclear factor kappa B in inflammatory bowel disease. Gut 1998;42:477–84. 10. Maconi G, Parente F, Bollani S, Imbesi V, Ardizzone S, Russo A, et al. Factors affecting splanchnic haemodynamics in Crohn's disease: a prospective controlled study using Doppler ultrasound. Gut 1998;43:645–50. 11. van Oostayen JA, Wasser MN, van Hogezand RA, Griffioen G, de Roos A. Activity of Crohn's disease assessed by measurement of superior mesenteric artery flow with Doppler US. Radiology 1994;193:551–4. 12. van Oostayen JA, Wasser MN, van Hogezand RA, Griffioen G, Biemond I, Lamers CB, et al. Doppler sonography evaluation of superior mesenteric artery flow to assess Crohn's disease activity: correlation with clinical evaluation Crohn's disease activity index, and alpha1-antitrypsin clearance in feces. AJR Am J Roentgenol 1997;168:429–33. 13. Ludwig D, Wiener S, Brüning A, Schwarting K, Jantschek G, Stange EF. Mesenteric blood flow is related to disease activity and risk of relapse in Crohn's disease: a prospective follow-up study. Am J Gastroenterol 1999;94:2942–50. 14. Paredes JM, Ripollés T, Cortés X, Martínez MJ, Barrachina M, Gómez F, et al. Abdominal sonographic changes after antibody to tumor necrosis factor (anti-TNF) alpha therapy in Crohn's Disease. Dig Dis Sci 2010;55:404–10. 15. van Oostayen JA, Wasser MN, Griffioen G, van Hogezand RA, Lamers CB, de Roos A. Diagnosis of Crohn's ileitis and monitoring of disease activity: value of Doppler ultrasound of superior mesenteric artery flow. Am J Gastroenterol 1998;93:88–91. 16. D'Haens GR, Panaccione R, Higgins PD, Vermeire S, Gassull M, Chowers Y, et al. The London Position Statement of the World Congress of Gastroenterology on Biological Therapy for IBD with the European Crohn's and Colitis Organization: when to start, when to stop, which drug to choose, and how to predict response? Am J Gastroenterol 2011;106:199–212. 17. Theodossiadis PG, Markomichelakis NN, Sfikakis PP. Tumor necrosis factor antagonists: preliminary evidence for an emerging approach in the treatment of ocular inflammation. Retina 2007;27(4):399–413. 18. Best WR, Becktel JM, Singleton JW, Kern Jr F. Development of a Crohn's disease activity index. Gastroenterology 1976;70:439–44. 19. Collins KA, Korcarz CE, Lang RM. Use of echocardiography for the phenotypic assessment of genetically altered mice. Physiol Genomics 2003;13:227–39. 20. Bonnin P, Pournaras JA, Lazrak Z, Cohen SY, Legargasson JF, Gaudric A, et al. Ultrasound assessment of early ocular vascular effects after intravitreal injection of bevacizumab (Avastin®) for neovascular age-related macular degeneration. Acta Ophthalmol 2010;88:641–5. 21. British standards institution: precision of test method (BS5497, part 1). London: BSI; 1979. 22. Debbabi H, Bonnin P, Ducluzeau PH, Lefthériotis G, Levy BI. Non-invasive assessment of endothelial function in the skin microcirculation. Am J Hypertens 2010;23:541–6. 23. Brown PW, Bargen JA, Weber HM. Chronic inflammatory lesions of the small intestine (regional enteritis). Am J Dig Dis 1934;1: 426–31. 24. Warren S, Sommers SC. Pathology of regional enteritis and ulcerative colitis. JAMA 1954;154:189–93. 25. Hultén L, Lindhagen S, Lundgren O, Fasth S, Ahrén C. Regional intestinal blood flow in ulcerative colitis and Crohn's disease. Gastroenterology 1977;72:388–96. 26. Meadows TR, Batsakis JG. Histopathological spectrum of regional enteritis. Arch Surg 1963;87:976–81. 27. Brahme F. Mesenteric angiography in regional enterocolitis. Radiology 1966;87:1037–42. 28. McGovern VJ, Goulston SJ. Crohn's disease of the colon. Gut 1968;9:164–76.

29. Perner A, Rask-Madsen J. Review article: the potential role of nitric oxide in chronic inflammatory bowel disorders. Aliment Pharmacol Ther 1999;13:135–44. 30. Kimura H, Miura S, Shigematsu T, Ohkubo N, Tsuzuki Y, Kurose I, et al. Increased nitric oxide production and inducible nitric oxide synthase activity in colonic mucosa of patients with active ulcerative colitis and Crohn's disease. Dig Dis Sci 1997;42:1047–54. 31. Millar TM, Stevens CR, Benjamin N, Eisenthal R, Harrison R, Blake DR. Xanthine oxidoreductase catalyses the reduction of nitrates and nitrite to nitric oxide under hypoxic conditions. FEBS Lett 1998;427:225–8. 32. Kanazawa S, Tsunoda T, Onuma E, Majima T, Kagiyama M, Kikuchi K. VEGF, basic-FGF, and TGF-beta in Crohn's disease and ulcerative colitis: a novel mechanism of chronic intestinal inflammation. Am J Gastroenterol 2001;96:822–8. 33. Nilsen EM, Johansen FE, Jahnsen FL, Lundin KE, Scholz T, Brandtzaeg P, et al. Cytokine profiles of cultured microvascular endothelial cells from the human intestine. Gut 1998;42:635–42. 34. Yoshida H, Yilmaz CE, Granger DN. Role of tumor necrosis factor-α in the extraintestinal thrombosis associated with colonic inflammation. Inflamm Bowel Dis 2011;17:2217–23. 35. Rogler G, Andus T. Cytokines in inflammatory bowel disease. World J Surg 1998;22:382–9. 36. Chiarantini E, Valanzano R, Liotta AA, Cellai AP, Fedi S, Ilari I, et al. Haemostatic abnormalities in inflammatory bowel disease. Thromb Res 1996;82:137–46. 37. Targan SR, Hanauer SB, van Deventer SJ, Mayer L, Present DH, Braakman T, et al. A short term study of chimeric monoclonal antibody cA2 to tumour necrosis factor alpha for Crohn's disease. Crohn's disease cA2 study group. N Engl J Med 1997;337:1029–35. 38. Powrie F, Leach MW, Mauze S, Menon S, Caddle LB, Coffman RL. Inhibition of Th1 response prevents inflammatory bowel disease in scid mice reconstituted with CD45RBhi CD4 + Tcells. Immunity 1994;1:553–62. 39. Haber HP, Busch A, Ziebach R, Stern M. Bowel wall thickness measured by ultrasound as a marker of Crohn's disease activity in children. Lancet 2000;355:1239–40. 40. Spalinger J, Patriquin H, Miron MC, Marx G, Herzog D, Dubois J, et al. Doppler US in patients with Crohn disease: vessel density in the diseased bowel reflects disease activity. Radiology 2000;217:787–91. 41. Borberg H, Tauchert M. Rheohaemapheresis of ophthalmological diseases and diseases of the microcirculation. Transfus Apher Sci 2006;34:41–9. 42. Cruz BA, Reis DD, Araujo CA. Minas Gerais Vasculitis Study Group. Refractory retinal vasculitis due to sarcoidosis successfully treated with infliximab. Rheumatol Int 2007;27:1181–3. 43. Theodossiadis PG, Liarakos VS, Sfikakis PP, Charonis A, Agrogiannis G, Kavantzas N, et al. Intravitreal administration of the anti-TNF monoclonal antibody infliximab in the rabbit. Graefes Arch Clin Exp Ophthalmol 2009;247:273–81. 44. Regatieri CV, Dreyfuss JL, Melo GB, Lavinsky D, Farah ME, Nader HB. Dual role of intravitreous infliximab in experimental choroidal neovascularization: effect on the expression of sulfated lycosaminoglycans. Invest Ophthalmol Vis Sci 2009;50:5487–94. 45. Koizumi K, Poulaki V, Doehmen S, Welsandt G, Radetzky S, Lappas A, et al. Contribution of TNF-alpha to leukocyte adhesion, vascular leakage, and apoptotic cell death in endotoxin-induced uveitis in vivo. Invest Ophthalmol Vis Sci 2003;44:2184–91. 46. Weber P, Husemann S, Vielhaber H, Zimmer KP, Nowak-Göttl U. Coagulation and fibrinolysis in children, adolescents, and young adults with inflammatory bowel disease. J Pediatr Gastroenterol Nutr 1999;28:418–22.


Hemodynamic effects of anti-TNFα in active Crohn’s disease