Oral and intravenous iron treatment in inflammatory bowel disease: Hematological response and quality of life improvement

Abstract

Background

The aim was to evaluate the efficacy and tolerance of oral and intravenous iron treatment in anemic inflammatory bowel disease (IBD) patients, considering both hematological and quality-of-life outcomes.

Methods

We performed a prospective multicenter study in IBD patients with iron deficiency anemia. Patients having hemoglobin >10 g/dL were prescribed oral ferrous sulfate. If hemoglobin <10 g/dL, intravenous (sucrose) iron was administered. Oral iron-intolerant patients were changed to intravenous treatment. Clinical (Truelove/Harvey–Bradshaw), hematological (response defined as hemoglobin normalization), and quality-of-life (shortened CCVEII-9 questionnaire) evaluations were performed at baseline and at 3 and 6 months.

Results

100 IBD patients (59 Crohn's disease, 41 ulcerative colitis) were included. Mean basal hemoglobin levels were 10.8 ± 1.3 g/dL (range, 6.6–12.9). Seventy-eight patients received oral treatment and 22 intravenous iron. Hemoglobin normalization was achieved in 86% of patients: 89% with oral, and 77% with intravenous iron. An IBD activity increase was not demonstrated in any patient. Four patients (5.1%) showed oral iron intolerance leading to discontinuation of treatment. No adverse events were reported for intravenous iron. Hemoglobin correlated with CCVEII-9 (P < 0.001). The CCVEII-9 score increased in patients who normalized hemoglobin levels in 3 months (from 58 ± 9 to 73 ± 10) or 6 months (54 ± 9, 68 ± 12, and 74 ± 10) (P < 0.001).

Conclusions

Oral iron treatment is effective and well tolerated in most IBD patients, and does not exacerbate the symptoms of the underlying IBD. Intravenous iron, on the other hand, is an effective and safe alternative treatment for iron deficiency anemia in more severely anemic or intolerant patients. Anemia correction with iron treatment is associated with a relevant improvement in the patients' quality of life.

anemia, ferritin, inflammatory bowel disease, iron

In addition to the intestinal signs and symptoms, inflammatory bowel disease (IBD) is associated with a number of extraintestinal manifestations. Furthermore, systemic manifestations of IBD may also include malnutrition and anemia.^1,–3 Compared with the average awareness of other extraintestinal complications such as arthritis or osteopathy, the topic of anemia in IBD has received little attention.^2,4 Although efficient therapeutic options have been developed for the treatment of IBD-associated anemia, treating this manifestation often has a low priority for gastroenterologists.^2,4 However, anemia is a clinically relevant condition that may affect quality of life, the ability to work,^5,6 and it is a comorbid condition that is associated with other diseases or even death.^1,7

It has been suggested that oral iron therapy is poorly tolerated in IBD, and may even increase IBD disease activity, possibly because of iron-mediated formation of toxic reactive oxygen species.^8,9 It has also been suggested that oral iron therapy is of poor efficacy in IBD patients.¹⁰ However, there are few published data to support such claims. Moreover, the most appropriate route for iron supplementation in patients with IBD has not been completely clarified. Only a few studies, generally including a small number of patients, have evaluated the efficacy of intravenous iron for the treatment of anemia in IBD patients. On the other hand, the repercussion of anemia on the quality of life of patients with IBD seems to be substantial, but the beneficial impact of iron therapy over quality of life has only been occasionally studied. Therefore, the aims of the present study were to evaluate the efficacy and tolerance of oral and intravenous iron treatment in anemic IBD patients, considering both hematological and quality-of-life outcomes.

Patients and Methods

Patient Population

This was a prospective multicenter study consecutively including Crohn's disease (CD) and ulcerative colitis (UC) outpatients with iron deficiency anemia. Informed consent was obtained from all patients and the study was approved by the Committee of Human Experimentation at all institutions. Diagnoses of CD and UC were established by standard clinical, radiological, histological, and endoscopic criteria.¹¹ The Montreal Classification of CD based on Age at Diagnosis (16 years or younger [A1], 17-40 years [A2], over 40 years [A3]), Location (terminal ileum [L1], colon [L2], ileocolon [L3], upper gastrointestinal [L4]), and Behavior [inflammatory/nonstricturing nonpenetrating [B1], stricturing [B2], penetrating [B3]), was used. For UC, the Montreal Classification based on the extent of the disease was used: proctitis, left-sided or distal (up to the splenic flexure) colitis, and extensive colitis (proximal to the splenic flexure).

Inclusion criteria was the presence of anemia, defined by hemoglobin <13 g/dL in males and <12 g/dL in females.¹² In addition, the patients had to have a biochemical profile suggestive of iron deficiency, defined by at least 1 of the following analytical findings: 1) low serum iron (<59 μg/dL); 2) low ferritin (<30 μg/L); 3) low transferrin saturation (<12%); or 4) high transferrin concentration (>400 mg/dL).

The following were considered exclusion criteria: 1) serious cardiopulmonary, hepatic, or renal disease; 2) clinical relapse of disease activity needing hospitalization within the last 3 months; 3) nonsteroidal antiinflammatory drug intake during the 3 months preceding enrolment; 4) low vitamin B₁₂ or folic acid levels; 5) administration during the last 3 months of any treatment for anemia (oral or intravenous iron, erythropoietin, or blood transfusion); 6) change in the dose of any of the IBD therapies or initiation of any new treatment within the previous 3 months.

Activity of Disease

Activity of IBD was evaluated by the Truelove modified index^13,14 in UC patients (including the following variables: number of stools, blood in stools, hemoglobin, albumin, fever, tachycardia, erythrocyte sedimentation rate, white blood cells, and potassium; inactive: <11 points), and by the Harvey–Bradshaw Index in CD patients¹⁵ (including the following variables: general well-being, abdominal pain, number of loose stools, abdominal mass, and complications; inactive: <5 points).

Laboratory Parameters

Full blood count, erythrocyte sedimentation rate, C-reactive protein, and α₁ glycoprotein (orosomucoid) were determined by routine assay at baseline and after 3 and 6 months of starting iron treatment. Iron levels, ferritin concentration, transferrin saturation, and transferrin concentration were also measured at those visits.

Iron Treatment

Patients having hemoglobin levels >10 g/dL were prescribed oral ferrous sulfate (Fero-gradumet) at a dose of 1 pill (525 mg, equivalent to 106 mg of elemental iron) per day, after the principal meal. Compliance with therapy was determined from the interrogatory. Oral iron-intolerant patients were changed to intravenous treatment (even though they had hemoglobin levels >10 g/dL). When hemoglobin concentration was <10 g/dL, intravenous iron was administered (Venofer, 5 mL = 100 mg iron sucrose) at a dose of 200 mg twice a week. Iron sucrose doses were diluted in 100 mL isotonic sodium chloride and then given as a continuous intravenous infusion for about 30 minutes. The intravenous iron requirement for patients was individually calculated according to Ganzoni's formula¹⁶: total iron deficit (mg) = [body weight (kg) × (target hemoglobin – actual hemoglobin g/dL) × 0.24] + 500.

Clinical and Hematological Controls

Clinical and hematological controls were performed at baseline and 3 months after starting iron treatment. The main analytical outcome in this study was complete response to iron treatment, which was defined as the complete normalization of hemoglobin concentration (≥13 g/dL in males, and ≥12 g/dL in females). In those patients with a complete response to oral iron at 3 months, it was discontinued. In those with a hemoglobin rise of ≥1 g/dL, but not achieving normal hemoglobin values, oral iron treatment was continued for an additional 3 months. Finally, patients with no response to oral iron (that is, hemoglobin change of <1 g/dL) were offered treatment with intravenous iron. At 3- and 6-month controls, adverse effects of iron treatment were assessed.

Quality-of-Life Questionnaire

Patients were asked to complete a quality-of-life questionnaire before starting iron treatment and at 3 months. In case of requiring additional iron treatment for another 3-month period, the quality-of-life questionnaire was repeated at 6 months. Patients were blinded to the result of hematological controls before the quality-of-life questionnaire was completed. A previously validated shortened questionnaire (CCVEII-9) was used.¹⁷ This 9-item questionnaire includes the most representative items from the Spanish version of the 36-item Inflammatory Bowel Disease Questionnaire (IBDQ-36), which has previously demonstrated its adequate validity and reliability.¹⁸ The 9 items were: nausea, delay social engagement, passing gas, bowel movements, abdominal cramps, unwell, fatigue, satisfied/happy, and energy. The correlation with the IBDQ-36 and the shortened CCVEII-9 has proved to be excellent.¹⁷ The CCVEII-9 score was obtained by adding the score of each item and transforming the result on a scale ranging from 0 to 100, where 100 corresponds to the best perception.

Statistical Analysis

For continuous variables, mean and standard deviations were calculated. For categorical variables, percentages (and 95% confidence intervals [CI]) were provided. Categorical variables were compared with the χ² test. Quantitative variables were compared with the Student's t-test for paired data. A P-value < 0.05 was considered statistically significant. Multiple linear regression analysis was performed to study the correlation between basal hemoglobin concentrations and basal quality-of-life score. The dependent variable was CCVEII-9 score, and independent variables were: age, sex, smoking habit (smokers or nonsmokers), IBD type (UC or CD), and hemoglobin concentrations (g/dL). Time-to-response curves were derived using the Kaplan–Meier method for both the oral and intravenous iron treatment. Stepwise multivariate analysis using the Cox model was used to investigate factors potentially associated with response to iron treatment. The dependent variable was response to iron (complete hemoglobin normalization) during follow-up, and independent variables were: age, sex, smoking habit, IBD type, and type of iron formulation (oral or intravenous). The global yield of the different iron-related parameters (iron, ferritin, transferrin saturation, and transferrin) or biological serological markers (erythrocyte sedimentation rate, C-reactive protein, and orosomucoid) to predict anemia response to intravenous iron treatment was calculated using the area under the ROC (receiver operating characteristic) curve.

Results

One-hundred IBD patients (59 with CD and 41 with UC) were prospectively included. Mean age was 40 ± 16 years, 35% were males, and 16% smokers. Mean time since the diagnosis of IBD was 4.5 ± 4 years (range, 1–15 years). Montreal Classification of CD patients was as follows: Age at Diagnosis (A1, 31%; A2, 44%; A3, 25%), Location (L1, 31%; L2, 23%; L3, 42%; L4, 4%), and Behavior (B1, 53%; B2, 25%; B3, 22%). Disease extension in UC patients was as follows: proctitis (18%), left-sided colitis (46%), and extensive colitis (36%). Fifty-eight percent of the patients were taking oral 5-aminosalicylates, 13% were using topical 5-aminosalicylates, 52% were receiving azathioprine or mercaptopurine, and 3% methotrexate.

Mean values of analytical parameters defining iron deficiency anemia at baseline are summarized as follows: hemoglobin levels 10.8 ± 1.3 g/dL (range, 6.6–12.9 g/dL; 11.3 g/dL for patients receiving oral iron, and 8.8 g/dL for those receiving intravenous iron), mean corpuscular volume 83.3 ± 8, transferrin saturation 11.8 ± 9, transferrin concentration 263 ± 74 mg/dL, iron 37.6 ± 21 μg/dL, and ferritin 43 ± 63 μg/L.

Seventy-eight patients (78%), having a mean hemoglobin of 11.3 ± 0.7 g/dL, received oral treatment, while 22 patients (22%), with a mean hemoglobin of 8.8 ± g/dL, were treated with intravenous iron. Compliance with treatment and follow-up was complete in all cases.

Response to iron treatment (that is, complete hemoglobin normalization) was obtained in 86% (95% CI, 79%–93%) of the patients: 89% (81%–96%) with oral iron, and 77% (55%–92%) with intravenous iron administration (Fig. 1). Response rates were similar for CD and for UC (83% and 90%). Eighty-three percent of the patients who responded to oral treatment responded in 3 months, and the 17% left did in 6 months. The percentages for intravenous iron response at 3 and 6 months were 71% and 29%, respectively.

Figure 1

Response to oral and intravenous iron treatment.

In the multivariate analysis using the Cox model, none of the studied variables were associated with response to iron treatment. None of the iron-related parameters (including iron, ferritin, transferrin saturation, and transferrin) were useful to predict anemia response to intravenous iron treatment, with areas under the ROC curves ranging from 0.45–0.55. In particular, the area under the ROC curve for transferrin concentration (which has previously been reported to predict response¹⁹) was 0.4 (95% CI from 0.14–0.67).

When disease activity indexes were compared before and after iron treatment administration, a statistically significant improvement was demonstrated, both in UC patients (mean Truelove index decreased from 10.8 ± 0.4 to 9.9 ± 0.8; P < 0.01) and in CD patients (mean Harvey–Bradshaw Index decreased from 0.9 ± 0.5 to 0.4 ± 0.3; P < 0.01); nevertheless, IBD was found to be inactive in all patients at inclusion. IBD activity increase was not demonstrated in any patient. These results were confirmed even after excluding the hemoglobin from the Truelove score in UC patients.

Four patients (5.1%, 2 with CD and 2 with UC) showed oral iron intolerance, including gastrointestinal side effects (nausea, abdominal pain, and constipation) which led to discontinuation of treatment. No adverse events were reported for intravenous iron treatment.

Mean basal CCVEII-9 score (before iron administration) was 56 ± 9, ranging from 32–82. Results of multiple linear regression showed a correlation between hemoglobin concentration and CCVEII-9 score (linear correlation coefficient, 0.51; determination coefficient, 0.26; P < 0.0001) (Fig. 2). The CCVEII-9 score increased (from 58 ± 9 to 73 ± 10 points) in patients who responded to treatment (normalizing hemoglobin levels) in 3 months (P < 0.001) (Fig. 3). Slightly slower increment was shown in patients who needed 6 months to respond (54 ± 9, 68 ± 12, and 74 ± 10 points at 0, 3, and 6 months; P < 0.001). However, the CCVEII-9 score did not change in nonresponders (52 ± 6, 55 ± 7, and 56 ± 7 points at 0, 3, and 6 months) (Fig. 3).

Figure 2

Correlation between hemoglobin concentrations and quality-of-life questionnaire (CCVEII-9) score.

Figure 3

Changes in quality-of-life questionnaire (CCVEII-9) score depending on the response to iron treatment.

Discussion

A frequent misconception is to believe that anemia is an uncommon process in IBD.³ On the contrary, anemia is common in these patients, although the reported prevalence of this condition has been markedly variable.^1,2,20 In a systematic review published in 2004, the prevalence of anemia in patients with IBD ranged from 9%–74%.²⁰ In a more recent systematic review,¹ which included 19 studies (mainly on CD) the figures ranged from 6%–74%. Finally, a comprehensive review just published in 2008 calculated a mean prevalence of anemia of 17%. Therefore, anemia may be considered the most common systemic complication of IBD.³

Iron deficiency is the main cause of anemia in IBD patients, as a consequence of dietary restrictions, malabsorption, and/or intestinal bleeding. The prevalence of iron deficiency in IBD patients ranges from 36%–90%,¹ with a mean value of 45%,³ which underlines the fact that this condition may me considered the rule more than the exception.

There may be a tendency to look upon anemia as an unavoidable accompaniment to IBD.^3,21 For example, anemia at the start in our patients was relatively mild, with mean hemoglobin values of 10.8 g/dL. Moreover, mean hemoglobin in patients who received oral iron treatment (which was the case for most of our patients) was 11.3 g/dL, that is, very close to normal values. Only in recent years has correction of anemia been highlighted as a specific therapeutic aim in IBD patients.²¹ It should not be assumed that some level of anemia is a normal finding in IBD patients, and consequently it need not be treated.^2,22 On the contrary, iron supplementation should be started as soon as anemia (hemoglobin <13 g/dL in males, and <12 g/dL in females) is detected.¹² Thus, the World Health Organization definition of anemia applies to patients with IBD.^2,4,22,23

Accordingly, the therapeutic objective of the treatment with oral iron should be to completely correct the anemia, and not only to partially increase the hemoglobin levels. Therefore, the main analytical outcome in our study was “complete” response to iron treatment. It is important to remember that, besides the change in hemoglobin levels, the primary therapeutic goal is to improve quality of life of the patients, and that the highest improvement in the quality of life is observed precisely when the hemoglobin levels increase from 11 to 13 g/dL.^4,24

Overall, 86% of our patients had a complete hematological response to iron treatment, and this figure was as high as 89% with oral iron treatment. Our results are in agreement with previous studies showing that oral iron is equally effective in IBD and in non-IBD patients.^25,26 Although it has been reported that oral supplementation obtains a slow response,²⁷ among those who responded in our study most of them (83%) did so relatively early (at 3 months), whereas only 17% needed 6 months of treatment to achieve normal hemoglobin concentrations. Nevertheless, even when iron treatment is correctly prescribed, the oral route has several limitations, 1 of the most relevant being that oral iron may be not well tolerated by patients, mainly due to gastrointestinal side effects.³ From a recent systematic review on the management of anemia in IBD, which included several studies prescribing oral iron,^{9,10,25,28,–31} intolerance rate was a common finding leading to discontinuation in up to 21% of the cases.¹ However, in our experience only 5% of the patients showed oral iron intolerance leading to discontinuation of treatment.

Perhaps the explanation for this high compliance and low incidence of digestive intolerance in our patients is the low dose of oral iron prescribed (just 1 pill/day of Fero-gradumet, the oral iron presentation prescribed in our study, which includes ≈100 mg of elemental iron). Although conventional wisdom “says” that up to 200 mg of elemental iron per day is required to correct iron deficiency anemia, this is probably incorrect.³² Since a maximum of 10–20 mg of oral iron can be absorbed per day, higher doses are questionable. In addition, there is no evidence to support high doses of iron in comparative trials.^33,–35 This makes sense from a physiologic standpoint, since it is well known that the iron absorptive process is very efficient yet saturable.³² In this respect, a single tablet of most of the ferrous salt preparations (for example, sulfate) provides more iron than the intestine is able to absorb in 1 day.^33,34 On the other hand, it has been suggested that nonabsorbed iron salts can be toxic to the intestinal mucosa,^{8,9,30,31,36,–39} and perhaps could activate the disease.^1,40 However, it has been recently shown that oral iron is equally well tolerated in IBD and in non-IBD patients, and that a tiny minority of IBD patients relapses in association with use of oral iron therapy,^6,25,26 which coincides with our experience (as no exacerbation of symptoms or increase in IBD activity was demonstrated after iron treatment). Nevertheless, as a control group without iron treatment was not included in our study, we cannot exclude that iron could slow the mucosal healing process and consequently prolong the time to remission. However, not only was no IBD activity increase demonstrated in any patient, but, in fact, a decrease in IBD activity was confirmed after iron supplementation.

The established indications for intravenous iron are: severe anemia (generally defined as hemoglobin <10 g/dL^1,20,41), need of quick recovery in mild anemia, intolerance to oral iron, and failure of oral iron.² The efficacy of intravenous iron for the treatment of iron deficiency anemia in the general population (without IBD) has been demonstrated in numerous studies.⁴² Although the experience with intravenous iron in IBD is more limited, it seems similarly encouraging. However, only few studies, generally including a small number of patients (compared with the large experience in non-IBD patients), have evaluated the efficacy of intravenous iron for the treatment of anemia in IBD (Table 1).^{6,19,29,30,43,–50} In our study, intravenous iron treatment was effective (completely normalized hemoglobin levels) in 77% of the patients. Moreover, among those who responded, most of them (71%) did so relatively early (at 3 months; however, as we tested the outcome the first time after 3 months, an earlier response may be missed). These efficacy results are in complete accordance with those previously reported in the literature and summarized in a recent review,³ where mean response of iron deficiency anemia to the treatment with intravenous iron formulation was of 73%—a considerably high figure. Several variables have been suggested to predict response to intravenous iron, as for example transferrin concentration.¹⁹ However, this could not be confirmed in our study, as none of the iron-related parameters or the biological serological markers were useful to predict therapeutic success.

Table 1

Studies Evaluating the Efficacy of Intravenous Iron for the Treatment of Anemia in Inflammatory Bowel Disease

Low-molecular-weight iron dextran, ferric gluconate, ferric carboxymaltose, and particularly iron sucrose have been shown to be safe,⁵¹ as the incidence of adverse effects—and in particular severe adverse effects—has been extremely low.^1,42,52 Thus, no allergic reactions have been reported with iron sucrose,^1,4 the formulation prescribed in our and in most studies. Accordingly, no adverse events were found for intravenous iron treatment in any of our patients.

The repercussion of anemia on the quality of life of general patients^53,54 and specifically in patients with IBD^4,6,20,22 is substantial. In our study, the mean basal CCVEII-9 score (before iron administration) was as low as 56 (this score ranges from 0–100). Furthermore, a significant correlation was found between hemoglobin levels and the quality-of-life questionnaire score. It has been suggested that “Gastroenterologists tend to tolerate reduced hemoglobin levels better than their patients.”² As it has accurately been noted by Gasche et al,^4,22 for a long time it was thought that the clinical symptoms of anemia occurred only when the hemoglobin level dropped abruptly. It had been argued that patients would adapt to low hemoglobin levels if anemia developed slowly. This has led to the concept of “asymptomatic” anemia. In truth, the term “asymptomatic” seems to reflect the fact that impairments in physical condition, quality of life, and cognitive function may pass unrecognized by both patients and their doctors. Therefore, the process of adaptation to chronic anemia would be, in fact, adaptation to lower quality of life.^4,22

Remarkably, the quality of life in IBD patients may be as low as in anemic patients with advanced cancer.⁵⁵ Moreover, chronic fatigue caused by anemia can debilitate, affect, and worry these patients as much as abdominal pain or diarrhea.⁴ Therefore, the beneficial impact over quality of life derived from anemia correction in IBD patients can be similar to the control of diarrhea.^4,6,46 In our study the CCVEII-9 score rapidly and markedly increased in patients who responded to iron treatment (e.g., normalized hemoglobin) in just 3 months. A slightly slower increment was shown in patients who needed 6 months to achieve hematological response. However, quality of life did not change in nonresponders.

In summary, the conclusions of our study are that hematological response—with complete normalization of hemoglobin concentration—to oral iron is achievable in the majority of IBD patients, with similar encouraging results to those previously described in other causes of anemia. Oral iron treatment is effective and well tolerated in most IBD patients, and does not exacerbate the symptoms of the underlying IBD. Intravenous iron (sucrose), on the other hand, is an effective and safe alternative treatment for iron deficiency anemia in more severely anemic or intolerant patients. Finally, anemia correction with iron treatment is associated with a relevant improvement in the patients' quality of life, which constitutes a strong argument for more widespread use of iron replacement in IBD anemic patients.

Author notes

Reprints: Javier P. Gisbert, MD, Playa de Mojácar 29, Urb. Bonanza, 28669 Boadilla del Monte, Madrid, Spain (e--mail: )

CIBEREHD is funded by the Instituto de Salud Carlos III. This study was not funded by any pharmaceutical company.