Predictors of Non-Response to Ongoing Conservative Therapy in Patients with Inflammatory Bowel Disease

Abstract

Crohn’s disease (CD) and ulcerative colitis (UC), which are part of the group of inflammatory bowel diseases (IBD), belong to the group of immune-mediated diseases and characterized by a chronic relapsing and chronically continuous course, which leads to serious exacerbations and consequences. Patients may undergo radical surgery, often for drug-resistant disease, and the costs associated with IBD are significant and rising. Over the past two decades, there has been a paradigm shift in the treatment of IBD. The therapeutic goal has shifted from eliminating symptoms alone to achieving combined (symptomatic and endoscopic) remission, which is associated with better outcomes, including a lower risk of relapse, need for corticosteroids, hospitalization, colectomy, and colorectal neoplasia. Despite all the successes, the trend of non-response to ongoing conservative therapy continues. Most studies monitor through endoscopic evaluation and a small number of laboratory tests. There is an important need to understand how noninvasive biomarkers can serve as accurate and reliable indicators for assessing inflammation and predictors of lack of response to therapy. The purpose of this publication is to provide evidence on the use of biomarkers to assess disease activity and predictors of non-response to therapy in patients with IBD.

1. Introduction

Inflammatory bowel diseases are a group of immune-inflammatory pathologies with features of chronicity of the pathological process (chronic recurrent, chronic continuous), with high risks of surgical interventions, malignancy and disability.

Predictors of the lack of a therapeutic response to ongoing conservative therapy may consist of several aspects, first of all, these are the features of the clinical course, endoscopic and histological activity, data from laboratory and imaging diagnostics.

A significant number of individuals with CD need surgical intervention, and 20% of patients with UC undergo proctocolectomy due to resistance to ongoing treatments. The limited research data on indicators for surgery treatment in IBD has led to the initiation of several studies aimed at evaluating risk factors that increase the likelihood of surgery in IBD patients.

In recent years, predictors of non-response to certain drug groups (5-aminosalicylic acid drugs, immunosuppressants, drugs from the genetically engineered therapy group, and so on) have been separately evaluated.

We would like to bring to your attention our mini-review on predictors of non-response to ongoing therapy, markers of aggressive course, as well as a new system for assessing the severity of inflammatory bowel disease, taking into account the marker “influence factor” on the aggravated course and the development of complications.

1.1 What do we know today?

The presence of bleeding, abscesses, fistulas, intestinal obstruction, peritonitis, cases of intestinal perforation and the development of toxic megacolon in this group of patients are urgent circumstances that dictate the need for urgent surgical intervention.

If toxic megacolon is suspected, clinical signs of systemic intoxication in combination with radiological signs (expansion of the diameter of the colon more than 6 cm) are of leading diagnostic importance.

Timely treatment of toxic megacolon begins with intravenous administration of glucocorticoids and antibiotics. However, if there is no significant clinical improvement within the first 3 days in terms of reduction in toxicity and the need for fluid and blood transfusions, resolution of clinical symptoms and normalization of laboratory values, the next step is to consider infliximab or cyclosporine as a drug.

If no improvement is observed after this period of treatment, or if the patient has already received biological therapy, surgical treatment should be immediately considered [1].

Fistulas commonly develop in CD and can be multifaceted, for example between the affected bowel and nearby tissues such as other bowel, bladder, genitals or perianal area. These fistulas can lead to the formation of subcutaneous abscesses. There are various treatment options for fistulas: immunomodulators, biological drugs, antibiotics. However, in cases of lack of therapeutic response to the therapy, the possibility of switching to surgery should be considered [2].

Abscesses usually require antibiotics with percutaneous drainage of abscesses >2 cm in diameter with optimization of Crohn’s therapy, but abscess recurrence is possible, so again surgical resection of the affected bowel is often required [2].

Partial or complete intestinal obstruction may be due to the formation of strictures due to severe damage to the lumen of the small intestine with ulceration and transmural inflammation. The stricture becomes symptomatic with superposition of inflammation and spasmodic process. Accordingly, the resulting obstruction can be eliminated by parenteral administration of glucocorticoids, however, in case of acute deterioration, prompt surgical intervention is also required. Fibrous strictures causing recurrent episodes of obstruction should be removed by elective surgery [2].

It is known that the use of biological drugs for treatment can reduce the need for surgery, but despite this, a significant proportion of patients with Crohn’s disease, from 60 to 80%, still require surgery. The situation is similar in ulcerative colitis: approximately 20% of patients with ulcerative colitis undergo surgery due to lack of response to previous drug therapy [3].

Some studies have cited high recurrence rates, need for hospitalization, extraintestinal manifestations, multiple hospitalizations for exacerbations, early age at diagnosis, smoking, extensive mucosal ulceration, high serum antibody titers, and NOD2 gene mutations, as predictors of the need for surgery in patients with IBD [4].

In a study by David Marti-Aguado and his colleagues, some important results were obtained [5]:

• The number of patients who did not respond to aminosalicylates was 28%.

• Among patients with a stepped therapy approach, 79% of cases required escalation of treatment within the first 5 years.

• Some predictors of a higher need for therapy have been identified: young age at diagnosis, extensive colitis, endoscopic severity, and the presence of extraintestinal manifestations.

• For those in the step-down group, the reason for escalation was steroid treatment in 76% of cases, steroid resistance in 13%, and extraintestinal manifestations in 5%.

• Increased need for treatment escalation was observed in 97% of patients receiving immunomodulators (IMMs), 57% receiving biological therapy, and 48% of patients receiving a combination of IMM and biological therapy.

• The probability of no need for escalation of therapy after 20 years of follow-up was 65%. Thus, in patients with stepwise treatment, escalation of therapy was observed 2 years after diagnosis, and in 79% of cases, after 5 years of observation, escalation of therapy had already been completed.

• Initial analysis identified several factors that were strongly associated with the need for more intensive treatment: age, dose of 5-ASA, use of a combination approach to treatment, presence of extensive colitis, endoscopic severity of the disease, number of hospitalizations and the occurrence of extraintestinal manifestations (EIM).

• For more potential variables, a multivariate Cox regression model was used to predict the likelihood of switching to escalation therapy: young age, extensive colitis, endoscopic Mayo subscore of 2 or higher, and the presence of extraintestinal manifestations (EIM) were key factors that indicated a higher risk of need escalation of treatment [4] (Table 1).

Characteristics	HR (95%CI)	p value
Age at diagnosis	2.31 (1.36–3.92)	<0.01
Disease extent
E1, proctitis	Reference
E2, left-sided colitis	1.24 (0.75–2.04)	0.40
E3, extensive colitis	1.65 (1.04–2.60)	0.03
Mayo endoscopic subscore	1.45 (1.02–2.06)	0.04
Baseline 5-ASA dose	0.92 (0.64–1.30)	0.63
Hospitalization	1.68 (0.65–4.31)	0.28
EIMs	2.04 (1.03–4.05)	0.04

Table 1.

Risk factors for non-response to 5-ASA within 20 years; multivariate cox regression model analysis (HR 95%CI) [4].

5-ASA: 5-aminosalicylate; CI: confidence interval; EIM: extraintestinal manifestations; HR: hazard ratio.

A multivariate logistic regression analysis was conducted to identify factors linked to various escalation strategies, each examined separately and adjusted for gender and disease duration. The utilization of immunomodulatory medications (IMM) and biological agents was found to be associated with several factors, including younger age, extensive colitis, the progression of disease extent, the severity of endoscopic disease, and extraintestinal manifestations (EIMs). Notably, the presence of EIMs at the time of diagnosis was linked to a heightened risk of both IMM (Odds Ratio [OR] 3.69, 95% Confidence Interval [CI] 1.78–7.66) and biological therapy (OR 5.10, 95% CI 2.40–10.85). On the other hand, the development of EIMs during the follow-up period (OR 2.15, 95% CI 1.20–3.86) and the presence of pseudopolyposis (OR 2.13, 95% CI 1.08–4.20) were exclusively associated with the use of IMM. Extensive colitis, prior exposure to IMM, and steroid resistance were significantly correlated with a heightened likelihood of undergoing surgery. The distribution and relationship of disease extension and extent progression with different step-up approaches are detailed in [4] Table 2. The authors conducted a subgroup analysis to compare the treatment requirements before and after the introduction of biologic therapies for UC following their approval by the European Medicines Agency in February 2006. The group from the post-biologic era (comprising 246 individuals) showed a notable decrease in the rates of colectomy (7% vs. 1%; p < 0.01) when compared to the pre-biologic era. Among the individuals who followed a step-up treatment approach, only 21% required treatment escalation within a 5-year follow-up period. Within this cohort, the authors conducted a subgroup analysis, which demonstrated the absence of any kind of significant divergences between patients who required treatment escalation within a relatively short period (≤5 years, n = 99) and those who had a longer time before requiring treatment escalation (>5 years, n = 27).

Disease extension	IMM (n = 122)	Biological therapy (n = 72)	Colectomy (n = 16)
E1 (proctitis)	Reference	Reference	n = 0 (0%)
E2 (left-sided)	n = 37 (30%); OR 2.07 (1.04–4.11); p = 0.04	n = 21 (29%); OR 1.77 (0.80–3.70); p = 0.17	Reference
E3 (extensive)	n = 58 (48%); OR 2.93 (1.45–5.89); p < 0.01	n = 33 (46%); OR 2.74 (1.24–6.09); p = 0.02	n = 15 (94%); OR 9.28 (1.14–75.31); p = 0.04
Extent progression	n = 35 (47%); OR 5.19 (2.68–10.04); p < 0.01	n = 23 (31%); OR 4.37 (2.15–8.89); p < 0.01	n = 1 (2%); OR 0.63 (0.07–5.76); p = 0.68

Table 2.

Multivariate logistic regression analysis showing association of disease extension and extent progression with different step-up approaches (expressed as: n (%); OR (95% CI); p value). Bold values indicate they are statistically significant [4].

CI: confidence interval; IMM: immunomodulators; OR: odds ratio.

Previously published studies have identified several risk factors for relapse and the need for colectomy. These include younger age, male gender, shorter disease duration, smoking, extensive colitis, and initial low hemoglobin levels (<10.5 g/dL) in patients primarily treated with 5-ASA medication [6, 7, 8, 9, 10]. When examining factors associated with the use of a step-up treatment approach, the authors’ data support the findings related to younger age and extensive colitis, consistent with previous research. However, in contrast to the relapse rate, the researchers did not observe any differences based on gender or smoking habits. It’s worth noting that a recent meta-analysis also found that smokers did not experience significantly fewer disease flares [11]. The association between age and non-response to 5-ASA aligns with other studies that have shown younger patients are more likely to require steroids or experience a shorter time to relapse [6, 12], ultimately leading to a lower age at which treatment escalation becomes necessary. Similarly, the relationship between the Mayo endoscopic subscore and the adoption of a step-up approach suggests that the severity of the initial disease might serve as a significant predictor of unfavorable treatment outcomes.

A study by Carsten Schmidt et al. showed that, this study identifies anemia and an early need for corticosteroids as predictors for a complicated course of disease in an inception cohort of patients with UC. By determining these parameters in routine clinical practice, our results may support the identification of patients who might benefit from early escalation of therapy [13] and a Risk Model was built to predict the complicated course of the disease (Table 3).

Time	Corticosteroids	Anemia	Probability of IS, %
After 6 months	No	No	10.5
	Yes	No	15.1
	No	Yes	13.7
	Yes	Yes	19.6
After 12 months	No	No	17.0
	Yes	No	24.0
	No	Yes	21.9
	Yes	Yes	30.7
After 24 months	No	No	23.0
	Yes	No	32.1
	No	Yes	29.5
	Yes	Yes	40.3

Table 3.

Risk model for prediction of complicated disease course [13].

In a study by Elena De Cristofaro and colleagues [14], other predictors for surgery were identified: low serum albumin levels and colonic dilatation in patients with acute severe ulcerative colitis (ASUC) who respond to intravenous corticosteroid therapy (IV CS) or treatment with infliximab (IFX). ASUC is characterized by severe flare-ups in approximately one-fourth of individuals with ulcerative colitis (UC), as defined by the Truelove and Witts criteria [15, 16, 17]. For those who do not respond to IV CS within 3-5 days, second-line treatments like cyclosporine or infliximab (IFX) are considered rescue options [18]. Although these interventions have significantly improved the outlook for patients with IV CS-refractory UC, particularly in terms of reducing the need for colectomy, more than one-fourth of patients still require urgent colectomy [19]. Multiple predictors of colectomy have been identified in this subset of ASUC patients. These factors encompass clinical characteristics (such as gender, stool frequency, prior exposure to biological therapies, and steroid dependence), biological indicators (including albumin, C-reactive protein [CRP], and hemoglobin levels), as well as endoscopic and radiographic findings (such as the presence of “mucosal islands” and colonic dilation) [19, 20, 21, 22]. Additionally, superinfections like Clostridium difficile and Cytomegalovirus have been recognized as additional prognostic factors [23, 24].

In patients with acute severe ulcerative colitis (ASUC) who show a positive response to intravenous corticosteroids (IV CS) or infliximab (IFX) during their initial hospitalization, the likelihood of requiring colectomy tends to rise if they experience a second hospitalization for ASUC [25]. Specifically, individuals who do not fully respond to steroid treatment have a 50% chance of needing colectomy within 1 year after being discharged, and this risk escalates to 70% within 5 years [26].

To address this concern, a recent scoring system was developed with the aim of distinguishing ASUC patients at low risk from those at high risk of undergoing colectomy within 1 year following hospitalization. This scoring system took into account previous exposure to TNF antagonists or thiopurines, the presence of Clostridium difficile infection, and serum levels of C-reactive protein (CRP) and albumin as predictive factors for colectomy. However, it’s worth noting that the aforementioned ranking system lacked radiological and endoscopic criteria, and the follow-up period after hospitalization was relatively short [27]. Consequently, the common knowledge of the factors predicting long-term colectomy in ASUC patients who have achieved a complete clinical response to medical therapy remains limited.

A rather interesting and timely review on predictors of non-response in ulcerative colitis was presented in 2015 by a group of researchers led by Cláudia Camila Dias [28]. Potential predictors of colectomy included such indicators as gender, degree of disease, smoking history, hospitalization and the need for treatment with corticosteroids at any stage of the course of ulcerative colitis (UC). Utilizing a random effects model, our analysis revealed that female patients exhibited a reduced risk of colectomy (Odds Ratio [OR] 0.78, 95% Confidence Interval [CI] 0.68 to 0.90). Importantly, there was no significant evidence of statistical heterogeneity (I2 = 49%) in aforementioned findings, as illustrated in [28] (Figure 1). The random effects model showed that patients with extensive disease had a higher risk of colectomy (OR 3.68 [95% CI 2.39, 5.69]); however, significant heterogeneity was found among them (I² = 67%) [28] (Figure 2).

Figure 1.

Gender as a predictor of colectomy [28].

Figure 2.

Disease extent as a predictor of colectomy [28].

Regarding smoking habits, the analysis using a random effects model indicated that individuals who smoked had a reduced risk of requiring colectomy (OR 0.55 [95% CI 0.33 to 0.91]). Importantly, there was no notable heterogeneity in these findings (I² = 0%) [28] (Figure 3).

Figure 3.

Smoking habits as a predictor of colectomy [28].

Patients who experienced hospitalization at any point in their disease course faced an elevated risk of undergoing colectomy, as indicated by the analysis (OR 4.13 [95% CI 3.23, 5.27]). Notably, the authors’ findings did not reveal any significant evidence of statistical heterogeneity (I² = 0%) [28] (Figure 4).

Figure 4.

Hospitalization as a predictor of colectomy [28].

An association between the use of corticosteroids and colectomy was found in four studies. Patients who took corticosteroids (oral or intravenous) at any time had a higher risk of colectomy (OR 2.10 [95% CI 1.05, 4.22]); however, significant heterogeneity was found (I² = 54%) [28] (Figure 5).

Figure 5.

Use of the corticosteroids as a predictor of colectomy [28].

The authors [28] performed a subgroup analysis focusing on two colectomy outcomes: due to lack of response to treatment/or within the first year after diagnosis and colectomy occurring throughout the disease. In this analysis, the factors influencing these results were: gender, degree of disease, smoking. Returning to earlier data that controlled for gender (Figure 1), extent of disease (Figure 2), and smoking habit (Figure 3), no significant differences in the likelihood of undergoing surgery, regardless of its nature, occurring within the first year after surgery were found diagnosis or at a later stage of the disease.

In evaluating C-reactive protein (CRP) as a potential prognostic factor for colectomy, the authors [28] encountered challenges from the seven studies reviewed that presented different risk estimates, differences in timing of CRP measurement, and different cutoff values ​​for risk assessment. These discrepancies prevented data from being combined for analysis.

None of three separate investigations [29, 30, 31, 32, 33, 34] has provided the median or mean CRP levels. Furthermore, none of these studies disclosed the specific method used for measuring CRP levels, which posed a challenge in arriving at any meaningful conclusions. With the exception of one study [29], all the others exhibited a noticeable trend where individuals with elevated CRP levels were more likely to undergo colectomy. Ho GT et al. [32] introduced a risk index to assist in identifying patients who did not respond to treatment, leading to colectomy. They observed that non-responders had higher CRP levels compared to responders (6.9 mg/L versus 3.9 mg/L, p 4 weeks after the first induction therapy (HR 1.15 [95% 0.82, 1.62]). In a study by patients requiring colectomy [33], those admitted with elevated CRP values had significantly higher levels than their counterparts (mean [standard deviation] of 116 [102] versus 43 [25] mg/L). Similarly, Lindgren et al. [34] identified CRP levels of ≥25 mg/L on the third day of hospitalization as a predictor of colectomy within the first 30 days of hospital stay. Colectomized patients also exhibited notably higher CRP levels on the third day after treatment compared to non-colectomized patients (36.3 mg/L versus 18.0 mg/L, with a p-value of 0.007). Furthermore, one-year post-diagnosis, patients with CRP levels ≥10 mg/L faced a significantly elevated risk of undergoing colectomy in the subsequent 4 years, with an odds ratio of 3.0 [95% confidence interval: 1.1, 7.8] [35]. Regarding the timing of infliximab induction, two studies reported an association between elevated CRP values and an increased risk of colectomy. One study observed this association at CRP levels ≥10 mg/L, with a hazard ratio of 5.11 [95% confidence interval: 1.77, 14.76] [31], while the other study noted it at CRP levels ≥5 mg/L, with a hazard ratio of 14.5 [95% confidence interval, 2.0, 108.6] [30].

There are several compelling reasons why prognostic factors hold significant importance in the context of UC management, for instance [36]:

1. Mucosal healing has become a crucial therapeutic goal in UC management [16, 37];

2. The introduction of drugs from the group of biological therapies, such as anti-tumor necrosis factor medications, which has opened up possibilities for promoting mucosal healing and sustaining prolonged clinical remission in UC patients [38]; and

3. The lead of early therapeutic interventions to more favorable outcomes in UC patients [16, 37].

Given these imperatives, it becomes paramount to develop a specialized clinical tool capable of assessing disabilities and identifying specific factors that can predict UC outcomes. Therefore, the clinical risk factors analyzed in aforementioned meta-analysis should be taken into account when creating new scoring systems or approaches for evaluating UC outcomes. The authors highlight certain characteristics that can aid healthcare professionals in identifying risk groups among UC patients. This represents a significant stride in establishing predictive factors for UC, demonstrating their practicality and influence on disease prognosis, particularly the risk of colectomy. Factors such as gender, disease extent, hospitalization history, the need for corticosteroid treatment, smoking habits, and CRP levels were found to be associated with the likelihood of colectomy. Although there are other markers that can assist clinicians in forecasting the course of UC, including genetic, serologic, and endoscopic findings, the authors’ meta-analysis has specifically focused on demographic and clinical characteristics. This choice was made due to the ease of assessing these factors at the time of diagnosis and during routine patient care. As for CRP, variations in risk estimation methods and inadequate information regarding measurement techniques precluded authors from combining the data for analysis.

Nonetheless, a notable trend emerged suggesting an increased risk of colectomy among patients with elevated CRP (C-reactive protein) levels. Among the findings, the most significant risk factors identified in this analysis were the presence of extensive disease, hospitalization and smoking habits. According to the research outcomes from subgroups within the patient population, the predictors for colectomy remained consistent whether it was due to nonresponse to treatment at the time of diagnosis or within the first year after diagnosis, or occurred during the course of the disease. Gender, disease extent, and smoking habits continued to show strong associations, emphasizing their importance in predictive models. The researchers attempted to assess factors like multiple hospitalizations per year and recurrent flares requiring steroid treatment as indicators of chronic morbidity. However, there was insufficient data in the studies for a comprehensive analysis of these factors. Additionally, the authors could not explore the roles of other clinical and demographic variables, such as age at diagnosis, either because the data were unavailable or because they could not be aggregated due to variations in measurement methods. Furthermore, the authors encountered limitations in the analysis of several clinical trials involving UC due to the absence of stratified outcomes, specifically regarding gender, disease extent, smoking habits, hospitalization rates, and the necessity for corticosteroids. The majority of the studies included in authors’ analysis were of good quality; however, they did identify shortcomings in how outcomes were adjusted for in four of them. The primary limitation researches encountered in their meta-analysis was the presence of heterogeneity among the studies, stemming from differences in their methodologies and approaches. To gauge the impact of this heterogeneity on authors’ results, they conducted a sensitivity analysis. The aforementioned analysis indicated that their findings remained relatively stable, suggesting that the results are robust and dependable. Nevertheless, it’s important to acknowledge these limitations when interpreting the results and drawing conclusions.

In summary, the mentioned-above review and meta-analysis have underscored the significance of certain clinical factors, including gender, disease extent, smoking habits, hospitalization, and the use of corticosteroids, in predicting the outcome of ulcerative colitis, particularly the risk of colectomy. Integrating these parameters into the development of predictive models for UC prognosis holds the potential to improve the clinical approach and enhance the outcomes for patients dealing with severe forms of the disease.

Another rather interesting study was carried out by Egyptian colleagues [39], which studied predictors of surgery in patients with inflammatory bowel disease.

This study encompassed 80 individuals diagnosed with IBD, falling within an age range of 34 to 65 years. Among them, 40 were male, and 29 were smokers. The average age at which the disease first manifested was approximately 25.81 ± 6.8 years, with half of the patients experiencing involvement solely in the colon, while about 33.75% had both colonic and small intestinal complications. Extraintestinal symptoms, such as uveitis, arthritis, and arthralgia, were present in 61.3% of cases, and 37.5% of patients had perianal issues, with 27.5% showing granulomas in their histopathological reports. The mean Clinical Disease Activity Index (CDAI) score for Crohn’s disease (CD) patients was 298.45 ± 83.3, and for those with ulcerative colitis (UC), the mean Mayo score was 8.65 ± 1.9 [39].

Within the group of 40 patients who underwent surgical procedures, 30% had UC, while 70% had CD. Among these surgical cases, the procedures included total colectomy (30%), fistulectomy (32.5%), resection anastomosis (17.5%), and abdominal abscess drainage (20%). Patients in the surgical group exhibited a higher prevalence of perianal disease and smoking. Additionally, the CDAI scores for CD patients and the Mayo scores for UC patients were notably higher in the surgical group (P value <0.001). Crohn’s disease and the presence of granulomas were more frequently observed in the surgical group (P value <0.05). Conversely, factors such as the age of disease onset, the location of the disease, and the presence of extraintestinal manifestations did not show any statistically significant differences when comparing both patient groups (P value >0.5) [39].

In the group of patients who underwent surgery, there were notable differences in certain medical markers compared to the non-surgical group. Specifically, surgical patients exhibited significantly higher levels of stool calprotectin (P value <0.001), CRP (P value <0.001), and erythrocyte sedimentation rate (ESR) (P value <0.001). Additionally, CD patients in the surgical group had lower serum albumin levels at the onset of the disease compared to their non-surgical counterparts (P value <0.001). However, there were no significant differences observed in terms of complete blood count (CBC) parameters between the two groups, as indicated in [39] (Table 4).

Variable	Group 1 Mean ± SD	Group 2 Mean ± SD	P value
Stool Calprotectin level (μg/g)	543.35 ± 210.1	342.12 ± 229.2	<0.001
CRP level (mg/dl)	67.57 ± 14.1	33.35 ± 10.6	<0.001
Albumin level at time of diagnosis (g/dl) for CD	3.10 ± 0.42	4.62 ± 0.39	<0.001
ESR(mm/h)	59.95 ± 5.1	30.75 ± 4.1	<0.001
Hemoglobin (g/dL)	10.14 ± 0.59	9.87 ± 0.63	0.058
Hematocrit %	35.9 ± 2.9	36.32 ± 2.9	0.52
White blood cells (×1000 cell/mm^3)	12.25 ± 0.9	14.3 ± 1.4	0.41
Platelet count (×1000 cell/mm^3)	519.2 ± 54.3	498.42 ± 44.3	0.064

Table 4.

Comparison between both groups as regard laboratory data [39].

SD standard deviation.

Cutoff values, sensitivity, and specificity of significant scores and laboratory parameters are attached in [39] Table 5.

Variables	Surgical TTT
	Cutoffs	Sensitivity	Specificity	Value
CDAI for CD	>287	90%	85%	0.01
Mayo score for UC	>8.5	90%	70%	0.01
Stool calprotectin level (μg/g)	>341.5	92.5%	78.5%	0.01
CRP (mg/dl)	>44	99%	87.5%	0.01
ESR (mm/h)	>55	92.5%	100%	0.01

Table 5.

Cutoff values of different scores and inflammatory markers in prediction of surgical intervention in IBD patients [39].

There are numerous viable treatments available for IBD, such as Crohn’s disease and ulcerative colitis. However, a portion of individuals either do not respond well to these treatments initially or eventually lose their positive response [40]. In such cases, when the disease becomes medically refractory, surgical procedures like intestinal resection or proctocolectomy may become necessary [16].

It is crucial to identify patients who are at a high likelihood of needing surgery for Inflammatory Bowel Disease (IBD). However, the ability to predict which patients will require surgical intervention is still not well-established. To address this issue, a comparative study involving two medical centers was conducted in Egypt, involving 80 patients with IBD. The primary goal was to identify the factors that increase the risk of surgical intervention in these IBD patients.

It is crucial to identify patients who are at a high likelihood of needing surgery for IBD. However, the ability to predict which patients will require surgical intervention is still not well-established. To address this issue, a comparative study involving two medical centers was conducted in Egypt, involving 80 patients with IBD. The primary goal was to identify the factors that increase the risk of surgical intervention in these IBD patients.

Regarding the age of disease onset, the study found that the average age was 25.81 years, with a standard deviation of 6.8 years. This finding aligns with a study by De Barros et al. [41], which also analyzed the epidemiological characteristics of 40 IBD patients undergoing biologic therapy. In their retrospective observational clinical study, they observed a similar trend where the average age of disease onset was around 25 years old.

In the presented study, the authors showed that, the most prevalent site of the disease in IBD patients was the colon. This finding is consistent with the study conducted by Lehtinen et al. [42], which also pointed to the colon as the most frequently affected location. Moreover, Aziz et al. [43] conducted a 5-year retrospective study to assess various clinical presentations of IBD and similarly found that the colon was the most commonly affected site.

This study’s results showed that there was no statistically significant difference in the age at which the disease began between patients who required surgery and those who underwent medical treatment (P value = 0.37). This aligns with the findings of Lee et al. [12], who reported that the rates of surgery were not significantly linked to the age of onset in Korean ulcerative colitis patients. Additionally, a larger North American registry study also demonstrated that the age at which UC begins does not correlate with the likelihood of surgical intervention [44].

In this investigation, authors observed that smoking was significantly more prevalent among patients in group 1 (P < 0.001), and this was identified as a predictor of surgical treatment. This observation is in line with the research conducted by Karczewski et al. [45], who explored the impact of tobacco use on the clinical course of IBD.

In this study, authors [39] observed that extraintestinal manifestations were present in 61.3% of the cases, which closely resembles the findings of the study by de Barros et al. [41], where extraintestinal manifestations were noted in 70% of the patients. However, in this research, despite being more prevalent in Crohn’s disease (CD) patients, there was no statistically significant difference between those who required surgery and those treated medically in terms of extraintestinal manifestations (P = 0.25). This contrasts with a retrospective analysis of endoscopic balloon dilatation of intestinal strictures in CD, which reported that the presence of extraintestinal manifestations and extensive preoperative disease were more likely to necessitate further dilatation or surgery following an initial dilatation [46]. It’s possible that these differences can be attributed to variations in the ethnic backgrounds of the study populations.

Authors identified statistically significant differences between patients who underwent surgery and those who received medical treatment based on the type of IBD. Specifically, CD patients were more prevalent than UC patients in group 1 (P = 0.02). Surgical patients were also more likely to have perianal disease (P < 0.01) and granulomas (P < 0.012). These findings are consistent with research conducted by Nasir et al. [47], who documented that perianal disease and the presence of the NOD2 genotype were the only independent factors associated with the need for surgery. Similarly, Agar et al. [48] revealed that the presence of complicated disease features at the time of diagnosis, such as intestinal stenosis and fistulas, were independent predictive risk factors for subsequent surgery in CD patients.

Authors [39] observed statistically significant differences in disease activity scores between the two groups. Specifically, for Crohn’s disease patients, the Crohn’s Disease Activity Index (CDAI) was significantly higher (P value <0.001), and for ulcerative colitis patients, the Mayo score was also significantly higher (P value <0.001) in group 1, which included patients requiring surgery. Notably, CDAI scores greater than 287 and Mayo scores exceeding 8.5 demonstrated high sensitivity and specificity in identifying patients who needed surgical treatment. These findings align with research conducted by Agar et al. [48], who reported that a CDAI score above 250 was an independent predictive risk factor for future surgery in CD patients.

Furthermore, this study found that the mean levels of C-reactive protein (CRP) were 50.46 ± 21.2 (mg/dl), indicating elevated CRP levels. Similarly, the mean erythrocyte sedimentation rate (ESR) and white blood cell (WBC) count were 45.35 ± 15.4 mm/h and 13.28 ± 10.9 × 1000 cell/mm³, respectively, indicating increased expression of these biomarkers. These observations are in agreement with the findings of Alper et al. [49], who conducted research correlating ESR and CRP levels with the diagnosis of IBD and with disease activity assessed through clinical, endoscopic, histological, and radiographic measures. They found that IBD patients consistently exhibited significantly higher ESR and leukocyte counts. Solem et al. [50] also supported these findings in their research, highlighting the elevated levels of these biomarkers in IBD patients.

Analysis in this study [39] revealed statistically significant differences between patients who underwent surgery and those treated medically regarding C-reactive protein (CRP) levels (P value <0.001) and erythrocyte sedimentation rate (ESR) (P value <0.001). In group 1, which included patients requiring surgery, both CRP and ESR levels were notably higher. Specifically, a CRP level above 44 mg/dL was identified as a predictor of the need for surgical treatment, with a high sensitivity of 99% and specificity of 87.5%. These findings are consistent with a prospective study conducted in Oxford that assessed severe ulcerative colitis patients, where CRP levels exceeding 45 mg/L were a strong indicator (85% certainty) of the requirement for colectomy [33]. Additionally, Henriksen et al. [51] investigated the predictive value of CRP levels in UC patients with extensive colitis and found that CRP levels above 23 mg/L at diagnosis were associated with an increased risk of surgery. Similarly, in patients with Crohn’s disease, a significant association was found between CRP levels at diagnosis and the risk of surgery, with CRP levels above 53 mg/L indicating an elevated risk [51].

Authors [39] also assessed stool calprotectin levels as a predictor of surgical treatment. A stool calprotectin level greater than 341.5 μg/g was identified as a predictor, with a high sensitivity of 92.5% and specificity of 78.5%. These findings align with the work of Kennedy et al. [52], who reported that an increased level of fecal calprotectin during the initial visit was associated with the subsequent progression of CD and the need for surgery. Calprotectin concentration in UC patients was also found to correlate with worse outcomes, including the need for surgery [53]. Theede et al. [54] demonstrated that in UC, a baseline stool calprotectin level exceeding 321 μg/g predicted unfavorable outcomes, including the requirement for surgery.

In conclusion, the Egyptian researchers found that several factors were linked to an increased risk of requiring surgical intervention in patients with IBD. These factors encompassed smoking habits, the presence of Crohn’s disease, perianal complications, the development of granulomas, elevated severity scores (such as CDAI and Mayo score), higher levels of stool calprotectin, increased C-reactive protein levels, and elevated erythrocyte sedimentation rates. Furthermore, they identified smoking, perianal disease, CDAI, Mayo score, stool calprotectin level, C-reactive protein level, and erythrocyte sedimentation rate as valuable predictors of the need for surgical treatment, as these factors exhibited strong statistical significance in patients who ultimately required surgical intervention [39].

Considering the variations observed in assessing predictors of non-response in patients with Inflammatory Bowel Disease (IBD), including the disease’s aggressiveness, we have reached the conclusion that the essential laboratory diagnostics used in the daily practice of specialists dealing with IBD should be reevaluated. While C-reactive protein and calprotectin are currently considered mandatory markers, we believe that it’s worth exploring additional laboratory parameters.

With this in mind, our study aimed to develop a monitoring system for evaluating the condition of IBD patients, taking into account a broader range of laboratory tests.

2. Materials and methods

In the period from August 2015 to June 2022, 388 patients with a diagnosis of IBD. The diagnosis of IBD was made following widely accepted criteria, in line with the guidelines provided by the European Crohn’s and Colitis Organization (ECCO) [55, 56]. To evaluate the disease’s clinical severity, we used established measures such as the Truelove-Witts Index (TW) and Mayo score for ulcerative colitis, and the Crohn’s Disease Activity Index (CDAI) and Harvey-Bradshaw index (HBI) for Crohn’s disease.

The study [57, 58] did not include patients with severe cardiac and nephrological pathology, as well as those who underwent surgery within the last 6 months or are in the stage of clinical and endoscopic remission. Thus, the study included 250 patients; of these, 112 (44.8%) had CD and 138 (55.2%) UC. The age of patients was from 18 to 60 years (39.4 ± 4.6), by gender: 129 (51.6%) women and 121 (48.4%) men. The duration of the disease before contacting a specialist doctor was 1.1-8.3 years (3.7 ± 1.2), patients were under dynamic observation from 6 to 48 months (18.4 ± 7.2). In addition to the general clinical examination, the following laboratory markers were obligatory determined: highly sensitive C-reactive protein (h/s-CRP), homocysteine, vitamin D, platelet level, eosinophilic catthione protein, α-TNF, IL-1β, IL-2, IL-4, IL-6, IL-8, IL-10, IL-18 in blood, calprotectin, lactoferrin and TM-2-pyruvate kinase (TM2-P) in feces, albumin level (micro- and macro) in urine. Patients, if necessary, underwent repeated studies (680 cases in total). In our statistical analysis of the collected data, we employed standard descriptive statistical methods, including calculating the arithmetic mean (M), standard deviation (σ), and standard error (m) for the measured characteristics. When comparing numerical values between different groups, we utilized Student’s t-coefficient, and we determined the probability of error (p). To explore the relationships between various indicators, we calculated the Pearson correlation coefficient (®).

3. Results of the study and their discussion

The results obtained during the study are presented in [57, 58] Table 6.

Number of studies laboratory indicator	CD (n = 325)	UC (n = 355)	Total (n = 680)
High Sensitivity CRP: (N 1-3 mg/L)	325 (100%)	355 (100%)	680 (100%)
before 3 mg/L	11 (3.4%)	17 (4.8%)	28 (4.1%)
from 3 to 10 mg/L	106 (32.6%)	98 (27.6%)	204 (30%)
from 10 to 20 mg/L	74 (22.7%)	127 (35.8%)	201 (29.5%)
from 20 to 30 mg/L	95(29.2%)	87(24.5%)	182 (26.7%)
more 30 mg/L	39 (12%)	26 (7.3%)	65 (9.6%)
Homocysteine: (N < 12 μmol/L)	325 (66.7%)	355 (69%)	680 (67.9%)
before 12 μmol/L	108 (33.3%)	110 (30.9%)	218 (32%)
from 12 to 16 μmol/L f	83 (25.5%)	97 (27.3%)	180 (26.5%)
from 16 to 20 μmol/L	52 (16%)	63 (17.7%)	115 (16.9%)
from 20 to 24 μmol/L	48 (14.7%)	49 (13.8%)	97 (14.3%)
more 24 μmol/L	34 (10.5%)	36 (10.1%)	70 (10.3%)
Vitamin D: (N > 30 ng/mL)	325 (100%)	355 (100%)	680 (100%)
30–20 ng/mL	108 (33.2%)	121 (34.1%)	75 (21.1%)
20–10 ng/mL	159 (48.9%)	174 (49%)	172 (48.6%)
less 10 ng/mL	58 (17.8%)	60 (16.9%)	106 (30.0%)
Platelets: (N 150,000–400,000/mm^3)	325 (76%)	355 (74.3%)	680 (72.0%)
before 400,000/mm^3	78 (24%)	91 (25.6%)	169 (24.8%)
from 400,000 to 440,000/mm^3	101 (31%)	124 (34.9%)	225 (33%)
from 440,000 to 480,000/mm^3	92 (28.3%)	101 (28.4%)	193 (28.4%)
from 480,000 to 520,000/mm^3	33 (10.2%)	28 (7.9%)	61 (9%)
more 520,000/mm^3	21 (6.5%)	11 (3.1%)	32 (4.7%)
Fecal calprotectin (N < 50 μg/g)	325 (96.6%)	355 (93.8%)	680 (95.1%)
before <50 μg/g	11 (3.4%)	22 (6.2%)	33 (4.8%)
from 50 to 150 μg/g	43 (13.2%)	29 (8.2%)	72 (10.6%)
from 150 to 250 μg/g	77 (23.7%)	59 (16.6%)	136 (20%)
from 250 to 500 μg/g	83 (25.5%)	95 (26.7%)	178 (26.2%)
more 500 μg/g	111 (34.1%)	150 (42.3%)	261 (38.4%)
Fecal lactoferrin (N < 7.25 μg/g)	325 (97.8%)	355 (98.3%)	680 (98%)
before <7.25 μg/g	7 (2.2%)	6 (1.7%)	13 (19.1%)
from 7.25 to 14.5 μg/g	46 (14.2%)	55 (15.5%)	101 (14.8%)
from 14.5 to 22 μg/g	87 (26.7%)	91 (25.6%)	178 (26.2%)
from 22 to 29.5 μg/g	90 (27.7%)	100 (28.2%)	190 (27.9%)
more 29.5 μg/g	95(29.2%)	103 (29%)	198 (29.1%)
Fecal TM-2-pyruvate kinase (N < 4 U/ml)	325 (98.2%)	355 (98.9%)	680 (98.5%)
before <4 U/ml	6 (1.8%)	4 (1.1%)	10 (1.5%)
from 4 to 6 U/ml	28 (8.61%)	34 (9.6%)	62 (9.1%)
from 6 to 10 U/ml	65 (20%)	72 (20.3%)	137 (20.1%)
from 10 to 20 U/ml	83 (25.5%)	99 (27.9%)	182 (27.8%)
more 20 U/ml	143 (44%)	146 (41.1%)	289 (42.5%)
a-TNF (N 4.6–12.4 pg/ml)	325 (97.2%)	355 (98.6%)	680 (97.9%)
from 0 to 4.6 pg/ml	9 (2.7%)	5 (1.4%)	14 (2%)
from 4.6 to 12.4 pg/ml	13 (4%)	19 (5.4%)	32 (4.7%)
from 12.4 to 25 pg/ml	70 (21.5%)	84 (23.6%)	154 (22.6%)
from 25 to 35 pg/ml	94 (28.9%)	102 (28.7%)	196 (28.8%)
бoлee 35 pg/ml	139 (42.7%)	145 (40.8%)	284 (41.8%)
IL-1β (N < 11 pq/ntl)	24 (7.4%)	33 (9.3%)	57 (8.4%)
IL-2 (N < 10 pq/ml)	19 (5.8%)	26 (7.3%)	45 (6.6%)
IL-4 (N < 4 pq/ml)	13 (4%)	17 (4.8%)	30 (4.4%)
IL-6 (N < 10 pq/ml)	37 (11.4%)	41 (11.6%)	78 (11.5%)
IL-8 (N < 10 pq/ml)	119 (36.6%)	137 (38.6%)	256 (37.6%)
IL-10 (N < 20 pq/ml)	82 (25.2%)	88 (24.8%)	170 (25%)
IL-18 (N < 261 pq/ml)	69 (21.2%)	73 (20.6%)	142 (20.9%)
Albumin in urine	152 (46.7%)	175 (49.3%)	327 (48%)
Microalbuminuria (less 30 mg/L) Macroalbuminuria (more than 30 mg/L)	108 (33.2%) 44 (13.5%)	124 (38.2%) 51 (14.4%)	232 (34.1%) 95 (13.9%)

Table 6.

Distribution of studied parameters in patients with ulcerative colitis and Crohn’s disease [57, 58].

We conducted separate analyses for the detection of each of these indicators in both the ulcerative colitis (UC) and Crohn’s disease (CD) groups, but no significant differences were observed (p > 0.05). Similarly, when we examined the results based on gender, there were no significant differences (p > 0.05).

To explore a potential relationship between endothelial dysfunction indicators and the patients’ conditions, as per recommended guidelines, we conducted computer-based data processing. The severity of changes in these studied endothelial dysfunction indicators was evaluated as a percentage relative to the permissible range of the normal values. This assessment considered both increases in the indicator beyond the upper limit of the norm and decreases below the lower limit. The results of this analysis are presented in [57, 58] Table 7.

Laboratory indicator/Severity	Norm	I degree (mild)	II degree (medium-severe)	III degree (severe)
h/s CRP	N	1.3 N	1.5 N	>1.5 N
Homocysteine	N	1.3 N	1.5 N	>1.5 N
Platelets	N	1.1 N	1.2 N	>1.2 N
Vitamin D	N	0.7 N	0.4 N	<0.4 N
Fecal calprotectin	N	2 N	3 N	>3 N
Albumin in urine	N	—	Micro-	Macro-
Points	0	1	2	3

Table 7.

Correlation of some indicators of endothelial dysfunction, inflammation and the severity of the clinical course in patients with inflammatory bowel disease [57, 58].

Our findings indicate that the severity of endothelial dysfunction, categorized between 4 to 6 points, corresponds to a mild clinical course of Inflammatory Bowel Disease (IBD). A score ranging from 6 to 9 points suggests a moderate level of severity, and if the score exceeds 9 points, it indicates a high degree of clinical disease severity. Clinical and endoscopic remission, on the other hand, corresponds to a score of 3 points or less. The degree of correlation between these indicators was found to be 0.863.

Currently, the diagnosis of IBD relies on protocols that encompass a comprehensive assessment, including clinical evaluation, instrumental procedures, laboratory tests, and pathological examinations. Instrumental studies primarily emphasize radiation-based imaging methods such as CT, MRI, and ultrasound, along with thorough endoscopic examinations involving multiple biopsies from at least five sections of the intestine. In laboratory diagnostics, in addition to routine assessments and the exclusion of opportunistic infections, the determination of fecal calprotectin (a marker of intestinal mucosal damage) in stool samples has gained importance. Recently, the assessment has extended to include fecal lactoferrin and the acute-phase inflammation marker, C-reactive protein (CRP), in blood tests.

While the current diagnostic approach offers clear advantages, it comes with two notable drawbacks: high costs associated with the study and delays in diagnosis due to the need for pathomorphological examination results. Another critical factor is patient compliance with diagnostic procedures like MRI, CT scans, and endoscopy. In contrast, laboratory diagnostics primarily rely on just two indicators: calprotectin levels in stool samples and CRP levels in blood tests.

The literature extensively discusses the potential for incorporating additional laboratory diagnostic methods [59, 60, 61, 62]. When choosing a range of laboratory tests, we proceeded from the fact that IBD is based on immunological, aseptic inflammation. It is known that any inflammation is a manifestation of dysfunction of endothelial cells, which are involved in all phases of acute and chronic inflammation [60]. Endothelial cells play a crucial role in creating a barrier between the bloodstream and body tissues. They also produce various factors that serve vital regulatory functions, contributing to the maintenance of overall bodily balance (homeostasis). It is widely acknowledged that endothelial dysfunction (ED) is a typical pathological process and a central factor in the development of numerous diseases and their associated complications, including IBD. In the context of IBD, the inflammatory process within the intestinal mucosa, particularly the infiltration of leukocytes, can damage the endothelial cells lining the blood vessels in the intestinal mucosa. This damage can lead to disturbances in microcirculation, the formation of microthrombi, and subsequent trophic changes. Currently, several parameters are examined to assess endothelial dysfunction. These include homocysteine levels, thrombocytosis, von Willebrand factor, endothelin, highly sensitive C-reactive protein (CRP), changes in lipid profiles, interleukin patterns, plasminogen activator inhibitor-1 (PAI-1), plasminogen activator inhibitor-2 (PAI-2), intercellular adhesion molecule-1 (ICAM-1), nitric oxide (NO), P- and E-selectins, among others. Additionally, recent research has identified vitamin D deficiency as a risk factor for the development of autoimmune conditions [60].

We have selected the most accessible, both in practical (availability/availability in the laboratory network) and in economic terms, studies of indicators of endothelial dysfunction and immuno-allergic changes. These indicators were: homocysteine, platelets, highly sensitive CRP, eosinophilic cationic protein and vitamin D in the blood, fecal calprotectin, lactoferrin and tumor 2-pyruvate kinase, as well as microalbumin in the urine [57, 58].

Our findings lead us to the conclusion that the severity of certain markers of endothelial dysfunction, as well as immune-allergic changes, directly correlates with the severity of a patient’s condition. However, it’s important to note that since there is no significant difference observed between the patient groups with ulcerative colitis (UC) and Crohn’s disease (CD), these changes exhibit a low level of specificity. Consequently, they are best applied in cases where a diagnosis has already been established.

The advantages of utilizing this approach include:

1. **Economic feasibility:** It is a cost-effective method.

2. **Speed of calculation:** Results can be obtained quickly.

3. **Ease of implementation for the patient:** The procedure is relatively simple for patients to undergo.

4. **Widespread availability in outpatient practice:** This method is readily accessible for use in outpatient healthcare settings.

4. Conclusion

Thus, in addition to the already existing predictors of lack of response to ongoing therapy, taking into account clinical, endoscopic, laboratory activity, we can add the extended laboratory diagnostics of IBD that we proposed, which allows, with an established diagnosis, not to resort to repeated expensive studies and obtain results in real time, allowing personalize the assessment of the severity of the patient’s condition. This allows you to create an individual system for monitoring the patient’s condition, make timely decisions on the need for additional studies and correction of therapy, which is especially important in outpatient practice in matters of identifying a lack of response to ongoing conservative treatment.

Conflict of interest

The author affirms that there are no conflicts of interest to disclose.