Association between iron status and inflammation markers with GDF15 and BMP6 in inflammatory bowel disease: a single center cross-sectional study

Abstract

Background

Inflammatory bowel diseases (IBD), including Crohn’s disease (CD) and ulcerative colitis (UC), are chronic disorders with relapsing and remitting disease courses. Several clinical indicators and biomarkers have been evaluated for their clinical utility in assessing disease prognosis and monitoring disease activity. This study evaluated the utility of growth differentiation factor-15 (GDF15) and bone morphogenetic protein-6 (BMP6) in inflammatory pathways and iron regulation in IBD.

Methods

Between September 2015 and March 2016, this single-center, cross-sectional study included 40 CD patients, 29 UC patients, and 70 healthy controls. The disease activity was measured using the Mayo score and the Crohn’s Disease Activity Index (CDAI). Serum iron, hematological indices, inflammatory markers, and the novel biomarkers BMP6 and GDF15 as determined by ELISA were evaluated.

Results

Both the CD and UC groups showed significantly lower serum iron (p < 0.001), reduced mean corpuscular hemoglobin concentration, and higher red cell distribution width than the control group. Inflammatory markers, including erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP), were elevated in IBD, particularly during active disease, with CRP increasing significantly in active UC compared to remission (p = 0.040). In contrast, BMP6 and GDF15 levels did not differ significantly between the disease groups or activity states. Nonetheless, GDF15 demonstrated moderate positive correlations with serum iron (r = 0.314), hemoglobin (r = 0.320), and mean corpuscular hemoglobin concentration (MCHC) (r = 0.344), suggesting a link to erythropoietic or iron-related processes rather than inflammation.

Conclusions

Conventional markers reliably confirmed iron deficiency and inflammation in IBD. BMP6 and GDF15 levels did not differ significantly across disease states, indicating limited utility as markers of disease activity. Notably, GDF15 was positively correlated with serum iron, hemoglobin, and MCHC, suggesting its association with iron metabolism rather than active inflammation. Larger longitudinal studies are warranted to validate this potential role.

Trial registration

Not applicable.

Graphical Abstract

Background

Inflammatory bowel diseases (IBDs), including Crohn’s disease (CD) and ulcerative colitis (UC), are one of the main contributors to the global burden of diseases, accounting for almost 5 million cases worldwide as of 2019 and have an increasing age-standardized prevalence in approximately three-fourths of the countries or territories in the world, with women having a higher number of cases, deaths, and disability-adjusted life-years [1]. These diseases are chronic inflammatory disorders of the gastrointestinal tract that are characterized by a relapsing-remitting disease course and debilitating clinical symptoms during the active disease periods are associated with significant morbidity [2, 3].

Besides the deteriorated overall health status during the disease course, IBDs could be further complicated by chronic disease anemia and other inflammatory conditions. Although anemia is usually due to iron deficiency in IBDs, it may also be associated with other causes, which are often difficult to evaluate using routine biochemical analyses due to the inflammatory conditions [4].

Ongoing research is elucidating methods for assessing iron metabolism and inflammatory conditions in IBDs, and the hepatocyte-secreted hormone hepcidin has emerged as critical in iron regulation. Moreover, hepcidin expression was also found to be related to the interactions of several other signals, and bone morphogenetic proteins (BMPs) were found to activate the signals for hepcidin transcription [5, 6]. Among the BMPs, BMP6 is the most functional regulator of hepcidin, and its expression is modulated by circulating iron levels [7]. Besides BMP6, growth-differentiation factor-15 (GDF15) has also been found to be overexpressed under stress conditions, such as inflammation, and has been shown to predict poor outcomes in various conditions, including IBDs [8–10]. However, there is a paucity of data on the roles of these novel biochemical markers and their utility during the active disease and remission phases of IBDs as indicators of disease course and/or as predictive tools to assess the negative consequences of these chronic disorders, including anemia and inflammation. Therefore, this study aimed to evaluate the status of BMP6 and GDF15 in CD and UC compared to routine follow-up markers and their association with iron metabolism and inflammatory markers in IBD.

Methods

This single-center, observational, cross-sectional study was conducted at the Gastroenterology Department of Dışkapı Yıldırım Beyazıt Training and Research Hospital between September 2015 and March 2016. Patients who met the inclusion criteria during the recruitment period were informed about the study and were included if they agreed to participate. This study was approved by the Clinical Research Ethics Committee of Dışkapı Yıldırım Beyazıt Training and Research Hospital (approval number: 28/21, date: 04.04.2016), all work was carried out in compliance with the Ethical Principles for Medical Research Involving Human Subjects outlined in the Helsinki Declaration in 1975 (revised in 2000).

The study group included individuals aged 18 to 80 years who had been diagnosed with Crohn’s disease (CD) or ulcerative colitis (UC) within the last three months. The control group consisted of healthy individuals with no acute or chronic diseases. Exclusion criteria were as follows: acute or chronic kidney disease, hepatic cirrhosis, cardiac failure, pregnancy, neurological or psychiatric disease, history of abdominal surgery, history of any malignancy, and abnormal thyroid hormone levels.

The primary outcome of this study was to evaluate the associations between GDF15 and BMP6 levels and serum iron and inflammation indicators, as well as clinical activity indices, in IBD.

Biochemical analyses

For GDF15 and BMP6 analyses, blood samples were collected early in the morning after a minimum of 10 h of fasting; the samples were centrifuged for 15 min at 1000 × g, and the serum was stored at -80 °C in aliquots until analysis. Samples were collected between September 2015 and March 2016 and were analyzed thereafter; prolonged storage at -80 °C is generally considered acceptable for cytokine and growth factor stability, though potential degradation over extended storage periods cannot be entirely excluded and should be acknowledged as a limitation. Serum BMP-6 and GDF15 levels were measured using a quantitative sandwich enzyme immunoassay technique (ELISA) according to the manufacturer’s instructions (CUSABIO, Wuhan, China). The intra-assay coefficient of variation (CV) was < 8%, and the inter-assay CV was < 10% for both GDF15 and BMP6 analyses. The limit of detection (LOD) for BMP6 was 15.6 pg/mL and for GDF15 was 7.8 pg/mL as per the manufacturer’s specifications; all samples in this study yielded values above the respective detection limits.

Disease activity

Disease activity was evaluated using the Mayo [11] and Crohn’s disease activity index (CDAI) scores [12]. A Mayo score ≤ 4 was classified as remission, and > 4 as active disease in UC. In CD, a score of less than 150 indicates remission, and a score of more than 150 indicates active disease.

Statistical analyses

Descriptive analyses were presented using frequency and percentage for categorical variables and using mean, standard deviation, median, 25th and 75th percentiles (interquartile range, IQR), and minimum and maximum (range) for continuous variables, after controlling for normal distribution using visual inspection with histograms and Q-Q plots and the analytical methods of the Kolmogorov-Smirnov and Shapiro-Wilk tests. Comparisons between two independent groups were performed using the chi-square test for categorical variables and using the t-test or the Mann-Whitney U test, depending on the normal distribution of variables. Comparisons between multiple independent groups were performed using the chi-square test for categorical variables and the Kruskal-Wallis test for non-normally distributed continuous variables. Post-hoc pairwise comparisons were performed using the Mann-Whitney U test with Bonferroni correction. Spearman’s rho was used for non-parametric correlation analyses. The correlation analyses should be interpreted as exploratory since no correction for multiple testing was applied. Statistical significance was set at p < 0.05. All statistical analyses were performed using the PASW 18.0 software (version 18.0; IBM Inc., Armonk, NY, USA). A priori sample size estimation was not performed and the sample size was determined by the number of eligible patients during the recruitment period.

Results

Forty patients with CD (Male/Female: 24/16), 29 patients with UC (Male/Female: 19/10), and 70 healthy controls (Male/Female: 23/47) with mean ages of 40.4 ± 12.6 years, 43.1 ± 14.4 years, and 39.6 ± 10.8 years, respectively, were included in the study. Among the patients with CD, 24 (60.0%) had ileocolonic, 12 (30.0%) had ileal, 2 (5.0%) had colonic, 1 (2.5%) had upper localized disease, and the localization was not documented for 1 (2.5%) patient. Regarding the disease behavior, 33 (82.5%) patients had inflammatory diseases, 2 (5%) had stricture-related, 2 (5%) had penetrating, and 3 (7.5%) had perianal diseases. Among patients with UC, 18 (62.1%) had extensive or pancolitis localization, and 11 (37.9%) had distal colitis. Among all IBD cases, thirty-six patients (52.2%) had active disease, and 33 (47.8%) had remission.

Serum iron levels were significantly lower in both CD (median: 54.5 µg/dL; p < 0.001) and UC (55 µg/dL; p < 0.001) groups than in the controls (82 µg/dL, p < 0.001). Mean corpuscular hemoglobin concentration (MCHC) and red cell distribution width (RDW) also differed significantly among groups, with the controls having significantly higher MCHC (both CD and UC vs. control p < 0.001) and lower RDW (CD vs. control p < 0.001, UC vs. control p = 0.001) than CD and UC. Inflammatory marker levels were elevated in the IBD groups (p < 0.001). Erythrocyte sedimentation rate (ESR) was significantly higher in UC (median: 18 mm/h) and CD (13.5 mm/h), and both were significantly elevated compared to the controls (10 mm/h) (CD vs. control p = 0.011, UC vs. control p < 0.001). Similarly, C-reactive protein (CRP) levels were significantly higher in patients with CD and UC than in controls (p < 0.001), although the difference between CD and UC was not statistically significant. No statistically significant differences were detected in BMP6 or GDF15 levels across the groups (Table 1).

Table 1.

Iron metabolism indicators and inflammation markers in inflammatory bowel disease and control groups

	Diagnosis
	Crohn’s Disease (n = 40)	Ulcerative Colitis (n = 29)	Control Group (n = 70)	p*	p** (post-hoc)
	Median (IQR)	Median (IQR)	Median (IQR)		CD-UC	CD-Control	UC-Control
Biochemistry
Serum Iron	54.5 (35.5–79.5)	55 (25–77)	82 (56–116)	< 0.001	0.367	0.001	< 0.001
Serum Iron Binding Capacity (TIBC)	292 (249.5-347.5)	331.5 (245–362)	284 (224–342)	0.327	0.655	0.231	0.232
Complete blood count (CBC)
Hemoglobin (HGB)	13.8 (12.65-15)	12.9 (11.85–14.4)	14.01 (13.2–15.1)	0.056	0.112	0.443	0.017
Hematocrit (HCT)	41.85 (39.8–45.4)	39.55 (37.25–44.4)	41.1 (38.8–43.7)	0.284	0.137	0.390	0.277
Mean corpuscular volume (MCV)	86.8 (80.3–91)	87 (80.5-89.25)	85.3 (81.1–88.1)	0.243	0.469	0.095	0.484
Mean corpuscular hemoglobin (MCH)	28.3 (25.7–29.7)	28.3 (25.25–29.25)	29.1 (27.6–29.9)	0.145	0.550	0.210	0.065
Mean corpuscular hemoglobin concentration (MCHC)	32.45 (31.85–32.95)	32.45 (31.7–33)	33.9 (33.3–34.5)	< 0.001	0.980	< 0.001	< 0.001
Red cell distribution width (RDW)	15.05 (13.65–16.2)	14.7 (13.5–17.8)	13.45 (13-14.1)	< 0.001	0.627	< 0.001	0.001
Inflammation markers
Sedimentation rate (ESR)	13.5 (8.5–20.5)	18 (15–34)	10 (4–15)	< 0.001	0.038	0.011	< 0.001
C-reactive protein (CRP)	5.53 (2.94–11.3)	7.23 (3.88–19.3)	3.39 (1.66–5.1)	< 0.001	0.132	0.002	< 0.001
Research biomarkers
BMP6 (pg/mL)	93.13 (32.56-123.83)	66.94 (29.2-218.38)	106.68 (41.11–191.2)	0.409	0.641	0.179	0.555
GDF15 (pg/mL)	22.24 (5.89–69.41)	7.72 (1.61–30.31)	8.96 (2.07–30.95)	0.106	0.062	0.069	0.754

CD Crohn’s Disease, UC Ulcerative Colitis, BMP6 Bone morphogenetic protein-6, GDF15 Growth differentiation factor-15

 Bold values indicate statistical significance (p<0.05)

*Overall group comparison p-value by Kruskal-Wallis test; **Pairwise post-hoc p-values by Mann-Whitney U test with Bonferroni correction. Overall p-values reflect the omnibus test across all groups; post-hoc p-values reflect specific pairwise comparisons and should be interpreted in the context of multiple testing correction

When grouped by disease activity status, significant differences were observed in iron metabolism and inflammatory markers among patients in remission, those with active disease, and controls. The serum iron was significantly higher in the control group (median 87 µg/dL) than in both the remission (68 µg/dL, p = 0.021) and active disease (51 µg/dL, p < 0.001) groups. MCHC levels were reduced in both remission and active IBD patients compared to controls (p < 0.001 for both), with no significant difference between the remission and active groups. RDW was elevated in the IBD group relative to that in the control group (remission, p < 0.001; active, p = 0.007). Inflammatory markers were also markedly increased in the active group compared to both the remission (ESR, p = 0.003; CRP, p = 0.014) and control groups (both p < 0.001). BMP6 and GDF15 levels did not show statistically significant differences between the groups, although GDF15 levels tended to be higher in patients with active disease (Table 2).

Table 2.

Iron metabolism indicators and inflammation markers in the active and remission phases of inflammatory bowel disease

	Disease Status
	Remission (n = 33)	Active (n = 36)	Control Group (n = 70)	p*	p** (post-hoc)
	Median (IQR)	Median (IQR)	Median (IQR)		Remission-Active	Remission-Control	Active-Control
Biochemistry
Serum Iron	68 (31–86)	51 (25–68)	82 (56–116)	< 0.001	0.071	0.021	< 0.001
Serum Iron Binding Capacity (TIBC)	285 (251–359)	318 (247–352)	284 (224–342)	0.342	0.928	0.235	0.231
Complete blood count (CBC)
Hemoglobin (HGB)	13.65 (12.6-15.35)	13.4 (11.95–14.75)	14.01 (13.2–15.1)	0.159	0.499	0.272	0.068
Hematocrit (HCT)	41.1 (38.9–46.8)	41 (37.9–44.9)	41.1 (38.8–43.7)	0.780	0.426	0.723	0.721
Mean corpuscular volume (MCV)	86.4 (82.4–91)	87.3 (80.25–89.4)	85.3 (81.1–88.1)	0.315	0.910	0.232	0.203
Mean corpuscular hemoglobin (MCH)	28.2 (27.2–29.7)	28.35 (25.35–29.45)	29.1 (27.6–29.9)	0.176	0.940	0.169	0.100
Mean corpuscular hemoglobin concentration (MCHC)	32.55 (31.85–32.95)	32.4 (31.7–33)	33.9 (33.3–34.5)	< 0.001	0.640	< 0.001	< 0.001
Red cell distribution width (RDW)	15.1 (13.9–16.3)	14.6 (13.3-17.15)	13.45 (13-14.1)	< 0.001	0.330	< 0.001	0.007
Inflammation markers
Sedimentation rate (ESR)	11 (8–18)	19.5 (15–32)	10 (4–15)	< 0.001	0.003	0.112	< 0.001
C-reactive protein (CRP)	4.95 (2.91–7.23)	8.4 (4.48–22.9)	3.39 (1.66–5.1)	< 0.001	0.014	0.031	< 0.001
Research biomarkers
BMP6 (pg/mL)	55.09 (28.08-118.62)	115.86 (31.26-170.91)	106.68 (41.11–191.2)	0.128	0.103	0.055	0.899
GDF15 (pg/mL)	10.12 (1.76–45.19)	17.1 (3.97–54.44)	8.96 (2.07–30.95)	0.464	0.620	0.626	0.198

BMP6 Bone morphogenetic protein-6, GDF15 Growth differentiation factor-15 

Bold values indicate statistical significance (p<0.05)

*Overall group comparison p-value by Kruskal-Wallis test; **Pairwise post-hoc p-values by Mann-Whitney U test with Bonferroni correction. Overall p-values reflect the omnibus test across all groups; post-hoc p-values reflect specific pairwise comparisons and should be interpreted in the context of multiple testing correction

In CD and UC patients, no statistically significant differences were observed in the serum iron or hematologic parameters when comparing the active and remission phases of the diseases. However, in patients with UC, CRP levels were significantly higher during active disease (median: 11.4 mg/L) than during remission (3.89 mg/L, p = 0.040). BMP6 showed higher concentrations in the active UC and CD phases than in remission, particularly in UC (119.84 pg/mL vs. 30.7 pg/mL), although this was not statistically significant. GDF15 also showed a non-significant increase during active disease in both the CD and UC subgroups (Table 3; Fig. 1).

Table 3.

Iron metabolism indicators and inflammation markers in disease and disease status subgroups

	Crohn’s Disease (n = 40)			Ulcerative Colitis (n = 29)
	Remission (n = 23)	Active (n = 17)	p	Remission (n = 10)	Active (n = 19)	p
	Median (IQR)	Median (IQR)		Median (IQR)	Median (IQR)
Biochemistry
Serum Iron	70 (36–88)	51 (35–65)	0.125	54 (26–79)	55 (25–68)	0.581
Serum Iron Binding Capacity (TIBC)	297 (245–359)	290 (252–326)	0.805	278.5 (251–391)	343 (224–362)	0.979
Complete blood count (CBC)
Hemoglobin (HGB)	13.8 (12.7–15.7)	13.3 (12.6–14.8)	0.451	12.6 (11.8–13.5)	13.5 (11.9–14.4)	0.605
Hematocrit (HCT)	42.5 (39.8–47.6)	41.05 (38.3-45.15)	0.359	38.9 (38.1–40.2)	41 (36.5–44.5)	0.749
Mean corpuscular volume (MCV)	86.6 (83-91.1)	87.9 (79.3–90.9)	0.497	83.3 (73.5–88)	87 (80.8–89.3)	0.460
Mean corpuscular hemoglobin (MCH)	28.2 (27.2–30)	28.9 (25.3–29.4)	0.630	27.5 (22.8–28.9)	28.3 (25.4–29.5)	0.445
Mean corpuscular hemoglobin concentration (MCHC)	32.7 (31.8–33.1)	32.2 (31.9–32.9)	0.529	32.4 (32.2–32.7)	32.5 (31.7–33)	0.805
Red cell distribution width (RDW)	15 (13.9–16.1)	15.3 (13.3–16.8)	0.622	16 (14.25–18.2)	14.5 (13.3–17.5)	0.339
Inflammation markers
Sedimentation rate (ESR)	11 (8–19)	18 (13–21)	0.103	15.5 (8–18)	21 (15–36)	0.050
C-reactive protein (CRP)	5.53 (2.78–9.96)	5.81 (3.44-20)	0.304	3.89 (2.91–7.23)	11.4 (6.18–22.9)	0.040
Research biomarkers
BMP6 (pg/mL)	67.53 (34.33-118.62)	100.25 (30.79-151.07)	0.268	30.7 (27.2–168)	119.84 (34.81-218.38)	0.192
GDF15 (pg/mL)	17.47 (3.82-255.16)	24.68 (6.4-54.44)	0.924	2.72 (0.16–10.12)	12.83 (3.42–30.31)	0.114

BMP6 Bone morphogenetic protein-6, GDF15 Growth differentiation factor-15

Bold values indicate statistical significance (p<0.05)

Fig. 1.

Box plots comparing GDF15 and BMP6 levels across disease groups and activity states. Inflammation markers (CRP, ESR), GDF15, and BMP6 are shown for Control, Remission, and Active Disease subgroups. CRP, C-reactive protein; BMP6, bone morphogenetic protein-6; GDF15, growth differentiation factor-15

Correlation analysis revealed that GDF15 showed consistent, moderate positive correlations with several iron-related and hematologic parameters across the entire cohort. Notably, GDF15 levels were positively correlated with serum iron (r = 0.314, p = 0.011), hemoglobin (r = 0.32, p = 0.009), and MCHC (r = 0.344, p = 0.005). These correlations were evident, particularly in the remission and active subgroups. Also, GDF15 was negatively correlated with TIBC (r=-0.341, p = 0.006), which was also observed in the active disease (r=-0.408, p = 0.023). In contrast, BMP6 did not demonstrate strong or significant correlations with most parameters. Within the Crohn’s disease subgroup, a moderate negative correlation was observed between BMP6 and mean corpuscular volume (MCV) (r=–0.372, p = 0.020) and MCH (r=–0.329, p = 0.041), while GDF15 maintained its positive association with hemoglobin and MCH (Table 4).

Table 4.

Correlations between BMP6 and GDF15 with iron metabolism indicators and inflammation markers

	Overall		Remission		Active
	BMP6 (pg/mL)	GDF15 (pg/mL)	BMP6 (pg/mL)	GDF15 (pg/mL)	BMP6 (pg/mL)	GDF15 (pg/mL)
	r (p)	r (p)	r (p)	r (p)	r (p)	r (p)
Inflammatory Bowel Disease
Biochemistry
Serum Iron	-0.014 (0.91)	0.314 (0.011)	0.115 (0.523)	0.363 (0.038)	-0.053 (0.773)	0.33 (0.065)
Serum Iron Binding Capacity (TIBC)	0.024 (0.851)	-0.341 (0.006)	0.002 (0.99)	-0.259 (0.145)	0.054 (0.771)	-0.408 (0.023)
Complete blood count (CBC)
Hemoglobin (HGB)	0.034 (0.785)	0.32 (0.009)	0.271 (0.134)	0.349 (0.05)	-0.129 (0.466)	0.323 (0.062)
Hematocrit (HCT)	-0.055 (0.669)	0.24 (0.056)	0.229 (0.215)	0.295 (0.107)	-0.298 (0.092)	0.214 (0.231)
Mean corpuscular volume (MCV)	-0.177 (0.158)	0.286 (0.021)	-0.209 (0.26)	0.222 (0.23)	-0.143 (0.421)	0.374 (0.029)
Mean corpuscular hemoglobin (MCH)	-0.075 (0.552)	0.335 (0.006)	-0.146 (0.435)	0.279 (0.129)	-0.004 (0.984)	0.393 (0.022)
Mean corpuscular hemoglobin concentration (MCHC)	0.159 (0.203)	0.344 (0.005)	0.097 (0.596)	0.293 (0.103)	0.24 (0.172)	0.405 (0.018)
Red cell distribution width (RDW)	-0.009 (0.944)	-0.138 (0.272)	-0.24 (0.194)	0.121 (0.516)	0.132 (0.456)	-0.322 (0.064)
Inflammation markers
Sedimentation rate (ESR)	0.043 (0.731)	-0.061 (0.628)	-0.066 (0.717)	-0.1 (0.578)	0.112 (0.543)	-0.049 (0.791)
C-reactive protein (CRP)	-0.096 (0.452)	-0.019 (0.882)	-0.133 (0.461)	-0.197 (0.271)	-0.157 (0.399)	0.145 (0.437)
Crohn’s Disease
Biochemistry
Serum Iron	-0.075 (0.646)	0.313 (0.049)	0.1 (0.648)	0.309 (0.152)	-0.259 (0.316)	0.386 (0.126)
Serum Iron Binding Capacity (TIBC)	0.074 (0.652)	-0.423 (0.007)	-0.061 (0.781)	-0.315 (0.143)	0.196 (0.451)	-0.643 (0.005)
Complete blood count (CBC)
Hemoglobin (HGB)	-0.08 (0.622)	0.332 (0.036)	0.212 (0.332)	0.276 (0.202)	-0.406 (0.106)	0.401 (0.11)
Hematocrit (HCT)	-0.207 (0.211)	0.228 (0.168)	0.151 (0.501)	0.199 (0.375)	-0.711 (0.002)	0.256 (0.338)
Mean corpuscular volume (MCV)	-0.372 (0.02)	0.284 (0.079)	-0.413 (0.056)	0.185 (0.409)	-0.329 (0.197)	0.448 (0.071)
Mean corpuscular hemoglobin (MCH)	-0.329 (0.041)	0.318 (0.048)	-0.383 (0.079)	0.222 (0.322)	-0.318 (0.213)	0.451 (0.069)
Mean corpuscular hemoglobin concentration (MCHC)	0.102 (0.529)	0.303 (0.057)	0.067 (0.762)	0.159 (0.467)	0.025 (0.925)	0.455 (0.067)
Red cell distribution width (RDW)	0.035 (0.831)	-0.188 (0.246)	-0.216 (0.323)	0.174 (0.426)	0.264 (0.307)	-0.563 (0.019)
Inflammation markers
Sedimentation rate (ESR)	-0.231 (0.151)	-0.195 (0.229)	-0.169 (0.441)	-0.114 (0.603)	-0.347 (0.172)	-0.406 (0.106)
C-reactive protein (CRP)	-0.315 (0.051)	-0.079 (0.632)	-0.211 (0.333)	-0.191 (0.382)	-0.568 (0.022)	0.041 (0.88)
Ulcerative Colitis
Biochemistry
Serum Iron	0.05 (0.812)	0.255 (0.218)	-0.018 (0.96)	0.511 (0.132)	0.075 (0.791)	0.232 (0.405)
Serum Iron Binding Capacity (TIBC)	-0.047 (0.829)	-0.219 (0.304)	0.358 (0.31)	-0.213 (0.555)	-0.117 (0.691)	-0.15 (0.609)
Complete blood count (CBC)
Hemoglobin (HGB)	0.18 (0.378)	0.256 (0.207)	0.083 (0.831)	0.452 (0.222)	0.129 (0.622)	0.222 (0.391)
Hematocrit (HCT)	0.155 (0.449)	0.213 (0.297)	0.192 (0.62)	0.408 (0.276)	0.054 (0.837)	0.189 (0.468)
Mean corpuscular volume (MCV)	0.082 (0.689)	0.271 (0.181)	-0.2 (0.606)	0.159 (0.683)	0.013 (0.959)	0.319 (0.212)
Mean corpuscular hemoglobin (MCH)	0.171 (0.404)	0.33 (0.1)	-0.083 (0.831)	0.285 (0.458)	0.143 (0.583)	0.336 (0.187)
Mean corpuscular hemoglobin concentration (MCHC)	0.177 (0.386)	0.401 (0.043)	-0.368 (0.33)	0.378 (0.316)	0.37 (0.144)	0.429 (0.086)
Red cell distribution width (RDW)	-0.104 (0.621)	-0.042 (0.841)	-0.286 (0.493)	0.095 (0.823)	0 (1)	-0.11 (0.673)
Inflammation markers
Sedimentation rate (ESR)	0.329 (0.108)	0.259 (0.212)	-0.024 (0.947)	-0.299 (0.402)	0.461 (0.084)	0.522 (0.046)
C-reactive protein (CRP)	0.116 (0.58)	0.137 (0.513)	-0.091 (0.803)	-0.377 (0.283)	0.157 (0.576)	0.236 (0.398)

CRP C-reactive protein, BMP6 Bone morphogenetic protein-6, GDF15 Growth differentiation factor-15 

Bold values indicate statistical significance (p<0.05)

Since the anemia might affect the association between GDF15 and BMP6 with the serum iron parameters, the correlation analyses were repeated according to the presence of anemia in the IBD patients. Accordingly, only GDF15 was found to have a strong negative correlation with TIBC in IBD patients with anemia (r=-0.915, p < 0.001) (Table 5).

Table 5.

Correlations between BMP6 and GDF15 with iron metabolism indicators and inflammation markers in patients with and without anemia

	IBD with anemia (N = 10)		IBD without anemia (N = 58)
	BMP6 (pg/mL)	GDF15 (pg/mL)	BMP6 (pg/mL)	GDF15 (pg/mL)
	r (p)	r (p)	r (p)	r (p)
Inflammatory Bowel Disease
Biochemistry
Serum Iron	-0.31 (0.383)	0.103 (0.776)	0.036 (0.794)	0.194 (0.16)
Serum Iron Binding Capacity (TIBC)	-0.20 (0.58)	-0.915 (< 0.001)	0.075 (0.591)	-0.159 (0.255)
Complete blood count (CBC)
Hematocrit (HCT)	-0.164 (0.651)	0.139 (0.701)	0.002 (0.987)	0.037 (0.791)
Mean corpuscular volume (MCV)	-0.139 (0.701)	0.285 (0.425)	-0.178 (0.193)	0.189 (0.167)
Mean corpuscular hemoglobin (MCH)	0.042 (0.907)	0.406 (0.244)	-0.058 (0.672)	0.235 (0.085)
Mean corpuscular hemoglobin concentration (MCHC)	-0.215 (0.551)	0.178 (0.623)	0.239 (0.075)	0.257 (0.056)
Red cell distribution width (RDW)	-0.03 (0.934)	-0.60 (0.067)	-0.041 (0.769)	0.062 (0.655)
Inflammation markers
Sedimentation rate (ESR)	0.419 (0.301)	0.204 (0.629)	-0.025 (0.854)	-0.068 (0.617)
C-reactive protein (CRP)	-0.127 (0.726)	0.079 (0.829)	-0.075 (0.594)	-0.117 (0.405)

IBD Inflammatory bowel disease, BMP6 Bone morphogenetic protein-6, GDF15 Growth differentiation factor-15 

Bold values indicate statistical significance (p<0.05)

Discussion

This study evaluated BMP6 and GDF15, which have been proposed as potential biomarkers with clinical utility in the assessment of iron metabolism and inflammation in patients with IBD. Our results confirmed that serum iron levels were significantly lower in IBD, along with other hematological parameters consistent with iron deficiency, in accordance with the known clinical course of these disorders. In addition, conventional inflammatory markers were significantly elevated in the active phases of CD and UC compared to other groups, reflecting the expected clinical picture. Importantly, neither BMP6 nor GDF15 levels differed significantly between CD and UC groups or between the active and remission phases of disease. These negative findings are scientifically valuable and should be explicitly acknowledged: BMP6 and GDF15 appear to have limited utility as markers of disease activity in IBD. Only a non-significant trend toward higher GDF15 levels was observed during active disease. Despite the lack of significant group differences, GDF15 showed positive correlations with serum iron, hemoglobin, and MCHC, suggesting a potential association with erythropoietic regulation or iron handling rather than with active inflammation. To exclude the possible interaction of anemia in the association between GDF15 and BMP6 with the serum iron metabolism parameters, we analyzed the correlations between these markers in the presence and absence of anemia in IBD, which showed that anemic status had no profound or negating effect on those correlations, highlighting that the correlations found in this study were driven primarily by the underlying IBD. This pattern was not observed for BMP6. Based on these results, GDF15 may be more consistently associated with systemic iron status than with disease activity across the different clinical states of IBD.

GDF15 has been reported to be a multifunctional biomarker that is responsive to cellular stress and may be a potential therapeutic target and diagnostic biomarker for different diseases. It belongs to the placental transforming growth factor-β (TGF-β) superfamily and is activated in response to stress conditions such as inflammation [13]. Previous evidence suggests that elevated levels of GDF15 are observed in inflammatory conditions including rheumatoid arthritis, psoriasis, and Behçet’s disease [14–16]. Given these associations, several studies have evaluated the role of GDF15 in IBD, which is also an inflammatory state. In one of these studies, Kucerka et al. [17] evaluated the role of GDF15 in severe IBD cases to assess disease severity and monitor disease activity, and reported that this biomarker was significantly associated with patient age and inflammatory markers in CD and UC. The authors stated that these inflammatory markers included fibrinogen and interleukin-6, which we did not evaluate in our study. In another study, Tonkic et al. [18] evaluated the association between GDF15 and IBD, and reported that serum GDF15 concentrations were significantly higher in IBD patients than in healthy controls, but no significant differences were observed between CD and UC. The authors also reported that GDF15 can predict CD and UC severity independent of age, sex, and CRP levels, and that GDF15 levels were higher in anemic patients with IBD, with a negative correlation between GDF15 and serum iron and hemoglobin levels. These results were contrary to our findings, which showed positive correlations between GDF15 and serum iron and hemoglobin. This discrepancy warrants careful consideration. The cohort of Tonkic et al. [18] likely included patients with established, long-standing IBD and a higher prevalence of anemia, whereas our cohort comprised newly diagnosed patients (within the preceding 3 months) with potentially less severe iron depletion. Furthermore, differences in medication exposure—particularly iron supplementation, which can suppress GDF15 production—may explain divergent correlation directions. In patients with severe iron deficiency anemia, GDF15 may be elevated as a stress signal while serum iron is suppressed, yielding a negative correlation; conversely, in early-stage IBD before chronic iron depletion is fully established, GDF15 may reflect compensatory erythropoietic activity that correlates positively with residual iron stores. A sensitivity analysis excluding patients on iron therapy would be needed to fully resolve this discrepancy, which was not feasible in the current study given the absence of treatment data. Moreover, given the moderate magnitude of the correlations found, the clinical significance should be considered cautiously and requires validation in larger, well-characterized cohorts. It is also worth noting that the observed associations between GDF15 and iron parameters may partly be driven by anemia status rather than underlying IBD activity per se; and our analyses to elucidate this possibility showed that the correlations presented in this study were driven by the IBD rather than the presence of anemia in our study group. However, given the small number of patients with anemia in our study, future studies should stratify analyses separately for anemic and non-anemic patients to clarify this relationship. In addition to its association with serum iron and inflammatory markers, GDF15 may also be related to other conditions in IBD. Yamamoto et al. [10] evaluated GDF15 levels and their association with sarcopenia in CD, assessed using the skeletal muscle index, and reported that GDF-15 was significantly higher with a low skeletal muscle index, suggesting GDF-15 could be an independent indicator of a low skeletal muscle index. This study revealed that GDF15 might play a role in a variety of metabolic interactions in IBD; however, in relation to iron metabolism and inflammation specifically, we could not demonstrate statistically significant correlations with inflammation. GDF15 was more strongly correlated with serum iron and related blood parameters, suggesting that it might be more valuable for assessing iron metabolism. Notably, the patients included in this study were diagnosed in the preceding 3 months before enrollment, emphasizing that the chronic inflammatory milieu typical of long-standing disease had not yet been fully established. In this early-stage cohort, the positive correlation of GDF15 with iron parameters is an intriguing finding. Beyond its role as an inflammatory marker, GDF15 is also known to be released during erythroid stress—elevated GDF15 levels have been reported in conditions with hyperproliferative erythropoiesis such as thalassemia [19]. In early IBD, GDF15 may rise as a compensatory erythropoietic signal before chronic inflammation fully suppresses bone marrow activity, which may explain the positive correlations found in this study and may yield novel hypotheses for further longitudinal investigations.

BMP6 plays an important role in iron regulation in organisms. This molecule exerts its effects on hepcidin, a protein that controls the outflow of iron from cells, and its deficiency is associated with increased intestinal iron absorption and its increased release from macrophages [20]. It is also associated with increased erythropoiesis in anemia [21]. Based on studies evaluating the role of BMP6 in iron metabolism, it was suggested to be beneficial in the treatment of anemia caused by hepcidin, and BMP6 agonists were considered in the treatment of iron overload disorders [7]. With regard to this, Wang et al. [22] evaluated the hepcidin inhibition and BMP6 levels in IBD in animal models, and reported that hepcidin upregulation during colitis was dependent on BMP6; however, it was not associated with increased BMP6 expression or activity, and interleukin-6 (IL-6) was required for increased hepcidin expression during colitis. This finding is critical for understanding why BMP6 levels did not differ significantly between groups in our study: in IBD, hepcidin is often driven directly by pro-inflammatory cytokines such as IL-6, potentially bypassing the classical BMP6 signaling pathway [22]. Therefore, BMP6 levels alone may not adequately reflect hepcidin activity or iron restriction in the context of active intestinal inflammation. In another recent study, Wozniak et al. [23] evaluated BMP6 levels in IBD patients and reported that CD patients with anemia had increased BMP6 levels compared with controls, but no significant differences were observed in UC. Our results partly support these previous studies in that we also could not find significant associations between BMP6 levels and iron metabolism and inflammation in IBD. The absence of IL-6 and hepcidin measurements in this study represents a critical limitation for the mechanistic interpretation of BMP6 findings, and future studies incorporating these measurements are needed to fully elucidate the BMP6-hepcidin axis in IBD.

Conclusions

This study is one of the few in the literature focused on the roles of GDF15 and BMP6 and their clinical utility in prognosis or monitoring in IBD. Although these biomarkers, particularly GDF15, demonstrated limited utility as markers of disease activity, they may be considered for monitoring iron metabolism and related hematological processes rather than for assessing inflammatory status. However, given the several limitations of this study, our results should be interpreted cautiously. First, this was a non-interventional observational study with a cross-sectional design, which limits causal or temporal inferences and is prone to selection bias. Any statements suggesting clinical or prognostic applications of GDF15 should be viewed as exploratory. Second, the single-center design limits the generalizability of findings, and the convenience sampling may introduce selection bias. Third, information on ongoing treatments—including iron supplementation, corticosteroids, immunomodulators, and biologics—was not collected. These therapies may substantially influence iron parameters, inflammatory markers, and GDF15 levels, and represent a potential source of confounding bias. Fourth, we did not analyze IL-6 and hepcidin levels, which are critical for mechanistic interpretation of BMP6 and GDF15 findings; specifically, hepcidin as the central regulator of iron is driven by IL-6 during inflammation, and its absence limits the interpretation of the BMP6-hepcidin axis. Fifth, the cohort is relatively small (40 CD, 29 UC), particularly when subdivided into active versus remission groups; non-significant trends for GDF15 and BMP6 may reflect insufficient statistical power (Type II error), and a formal power analysis was not performed. Sixth, the lack of disease-duration stratification and the early-stage nature of the cohort (diagnosed within the preceding 3 months) limit the applicability of findings to long-standing IBD. Finally, the sex distribution differed significantly between controls (23 M/47F) and CD patients (24 M/16F), which may have affected iron markers and should be acknowledged as a potential confounder. Despite these limitations, our results contribute to the growing evidence on this topic given the paucity of studies focused on GDF15 and BMP6 in IBD.

Abbreviations

BMP6

Bone morphogenetic protein-6

CDAI

Crohn's Disease Activity Index

CD

Crohn’s disease

CRP

C-reactive protein

CV

Coefficient of variation

ESR

Erythrocyte sedimentation rate

GDF15

Growth differentiation factor-15

HCT

Hematocrit

HGB

Hemoglobin

IBD

Inflammatory bowel diseases

IQR

Interquartile range

MCH

Mean corpuscular hemoglobin

MCHC

Mean corpuscular hemoglobin concentration

MCV

Mean corpuscular volume

PTH

Parathyroid hormone

PT/INR

Prothrombin time/international normalized ratio

RDW

Red cell distribution width

TIBC

Serum iron binding capacity

TGF-β

Placental transforming growth factor-β

UC

Ulcerative colitis

Authors’ contributions

Conceptualization and design: Y.Ü., A.Y.Ü; Data curation: Y.Ü., S.Y.; Formal analysis: Y.Ü., S.Y., Y.C; Investigation: Y.Ü., S.Y., F.U.; Methodology: Y.Ü., İ.Y., A.Y.Ü.; Project administration: Y.Ü.; Resources: Y.Ü., Y.C., İ.Y.; Supervision: H.E., Y.Ü., Validation: Y.Ü., A.Y.Ü; Visualization: A.Y.Ü, Y.C; Writing – original draft: Y.Ü.; Writing – review & editing: All authors. Each author has participated sufficiently in the work to take public responsibility for the content. All authors state that they have approved the final draft submitted.

Funding

This study was granted (supported) by Dışkapı Yıldırım Beyazıt Education and Research Hospital Scientific Research Projects Administration’s 10.15. 2015/55 resolution.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Clinical Research Ethics Committee of Dışkapı Yıldırım Beyazıt Training and Research Hospital (approval number: 28/21, date: 04.04.2016), all work was carried out in compliance with the Ethical Principles for Medical Research Involving Human Subjects outlined in the Helsinki Declaration in 1975 (revised in 2000). All participants in the study provided informed consent for participation in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.