Transient Impact of Dysglycemia on Sputum Conversion among Smear-Positive Tuberculosis Patients in a Tertiary Care Facility in Ghana

Abstract

BACKGROUND

Apart from increasing the risk of tuberculosis (TB), diabetes may be associated with more severe disease and lower rates of sputum conversion among TB patients.

METHODS

We conducted a baseline cross-sectional study with a longitudinal follow-up of newly diagnosed smear-positive TB patients for 6 months. Sputum conversion rates between those with dysglycemia and those without were compared at 2 months (end of the intensive phase) and 6 months (end of the treatment). Descriptive statistics and logistic regression were computed to assess factors associated with dysglycemia as well as sputum conversion.

RESULTS

A significantly higher proportion of normoglycemic patients had negative sputum compared with those with dysglycemia (83% vs 67%, P-value < .05) at 2 months but not at 6 months (87% vs 77%, P-value > .05). After controlling for age group and adjusting for other covariates, patients with dysglycemia were 66% less likely to convert sputum than those with normoglycemia. Females were at least 7 times more likely than males and those with high waist-to-hip ratio (WHR) of 88% were less likely compared with those with low WHR for sputum conversion at 2 months, respectively. At 6 months, females (compared with males) and those with high WHR (compared with those with normal WHR) were at over 9 times increased odds and 89% less likely for sputum conversion, respectively.

CONCLUSION

A significantly lower proportion of smear-positive TB patients with dysglycemia converted to smear negative after 2 months of treatment but not at the end of the treatment, thus suggesting a transient impact of dysglycemia on sputum conversion.

Keywords

Tuberculosis smear Positive dysglycemia sputum Conversion transient.

1. Introduction

Worldwide, tuberculosis (TB) infections are still high despite the many strategies to curtail it. In 2016, there were an estimated 10.4 million new TB cases worldwide with about 1.7 million people dying from it, making it the topmost infectious killer.¹ With the high global burden of diabetes and an increasing trend, especially in type 2 diabetes, the recognized reciprocal negative impact of TB and diabetes on each other is likely to worsen.^2–4 Diabetes represents a significant population risk for TB infections of ∼1.5 to 7.8 times^5–8 and may also be related to the development of multidrug resistant TB with an odds ratio (OR) of 2.1.⁹ Diabetes may be associated with more severe disease, higher rates of reactivation of old TB foci, more cavitations, and higher risk of death among TB patients.^6,9–11 The relationship between rates of sputum conversion among TB patients with diabetes or dysglycemia appears inconsistent.^9,12–16 Some studies have suggested a reduced rate at 2 months (end of the intensive phase) and 6 months (end of the treatment)^9,12 while others have shown no relationship between diabetes and sputum conversion rate at the end of the secondmonth.^14,17 More data are needed to help ascertain the impact of dysglycemia on the sputum conversion rate among TB patients. The additional information would help determine if adjustments must be made to the current treatment regime. Thus, the present study ascertained the impact of dysglycemia on sputum conversion among smear-positive TB patients in a tertiary care setting.

2. Materials and Methods

2.1 Study Design and Site

The study adopted a cross-sectional baseline assessment with a longitudinal follow-up for 6 months at the outpatient referral chest clinic of the Korle-Bu Teaching Hospital, which is a tertiary care facility in Ghana.

2.2 Participants and Sampling

We consecutively enrolled patients who were either first diagnosed at or referred to the Korle-Bu chest clinic as newly diagnosed smear-positive TB patients. Included patients were those aged 18 years or older, had no previous TB treatment, and who gave informed written consent. Patients who were diagnosed with smear-negative TB or extrapulmonary TB, those who had previously been treated for TB, or refused to give informed consent were excluded from the study.

Differences in sputum culture conversion rates comparing pulmonary TB patients with and without diabetes mellitus (DM) from previous studies have varied between zero and 15%.^18–22 Assuming a 15% difference in sputum conversion rates between pulmonary TB patients with or without DM, 134 smear-positive TB patients would be adequate to detect a difference in sputum conversion based on a power of 80% and a 2-sided confidence interval (CI) of 90%. Sampling was done on weekdays for 16 months until a sample size of 200 was obtained of which 171 had complete data. Anti-TB treatment regimen used included a combination of rifampicin (R), isoniazid (H), pyrazinamide (Z) and ethambutol (E) for the intensive phase (first 2 months) while only rifampicin (R) and isoniazid (H) were used in the continuation phase (next 4 months). None of the recruited participants admitted to being a known diabetic prior to the study.

The center employs the directly observed therapy strategy (DOTS) to improve compliance to treatment. Moreover, the TB control program employs treatment supporters to help patients adhere to their treatment in the community. They pay visits and assess compliance to treatment; challenges identified are remedied. This ensures a high rate of compliance to medications and successful treatment outcomes.

2.3 Measurements

Patients enrolled were given a data abstraction tool to capture data concerning their sociodemographic and anthropometric characteristics as well as medical history.

To assess their glycemia status, a 75 g oral glucose tolerance test (OGTT) was administered to all patients. A 10 mL fasting blood sample was taken into fluoridated blood sample tubes (kept on ice and centrifuged within 15 min of blood draw), ethylene-diamine-tetra-acetic acid (EDTA) tubes, and plain tubes.²³ Patients then received 75 g of glucose in 250 mL of water, and after–2-h blood sample was taken into fluoridated sample tubes and processed similarly as the fasting sample. Plasma glucose was determined using glucose oxidase commercial reagent kits and controls (Diasys GmBH, Germany). Diabetes was diagnosed when the fasting plasma glucose (FPG) and the 2-h postprandial (2HPP) blood glucose level were >7 mmol/L and >11.1 mmol/L, respectively, or on regular medication for diabetes. FPG and 2HPP values below 6.1 mmol/L and 7.8 mmol/L respectively are normal. Diagnoses of impaired or borderline fasting and glucose tolerance were made when FPG and 2HPP values were 6.1 to 6.9 mmol/L and 7.8 to 11 mmol/L, respectively.²⁴ None of the participants was known to have diabetes prior to enrollment. Patients found to have diabetes were referred to receive appropriate care.

The presence of Mycobacterium tuberculosis (MTB) was tested using the Cepheid GeneXpert system, a rapid, nucleic acid amplification test (NAAT) through polymerase chain reaction to confirm the diagnosis. Sputum smear microscopy using Ziehl–Neelsen staining was however used to ascertain sputum conversion at 2 months (end of the intensive phase) and 6 months (end of the treatment). The sputum conversion rate was determined as the percent of enrolled smear-positive pulmonary TB cases that converted to smear-negative status at 2 months and 6 months. For the purposes of analyses, participants with “very low” and “low” detected MTB and those with “medium” and “high” detected MTB by GeneXpert were recategorized as low and high loads, respectively.

Body mass index (BMI) was calculated and categorized as obese, overweight, normal, and underweight based on the cut-offs of 30.0 or more, 25 to 29.9, 18.5 to 24.9, and <18.5 (kg/m²), respectively.²⁵

2.4 Statistical Analysis

Data were analyzed with the statistical software Stata version 15 after initial capture with Microsoft Excel 2010. Analyses for the various characteristics were performed for patients with complete data. For the purposes of analyses, patients were categorized as either dysglycemia or abnormal (impaired glucose tolerance and diabetes) or normal using 2HPP values. Sociodemographic, anthropometric, and clinical variables were summarized and compared between these two groups using the χ² test. A similar analysis using FPG was not performed due to the very few numbers involved.

Two approaches of statistical analysis were performed involving the descriptive χ² test and logistic regression analysis. Descriptive cross-tabulation analysis was performed to assess covariates significantly associated with 2HPP using the χ² test. Due to the dummy nature of our study outcomes, a binary logistic regression model was used to test the effect of sociodemographic, clinical, and lifestyle factors on 2HPP and sputum conversion separately. Logistic regression analysis was conducted to estimate factors significantly influencing 2HPP. Age group as the only significant factor influencing 2HPP was controlled and other covariates were adjusted in further analysis to quantify the impact of 2HPP on sputum conversion for 2 months and 6 months separately by adopting logistic regression. The statistical tests were set at the 5% significance level.

3. Results

3.1 Participant Characteristics

The ages of the participants were evenly distributed across the age groups with a mean age of 38.2 ± 13.6 years (range 18–75 years). The majority of participants (74.8%) were educated up to senior high/ordinary level (SHS/O level) and 68.2% were underweight. Males represented 78.9% of the study participants. The test of association suggests that 2HPP is associated with only age group and educational level (P-value < .05). The result shows that generally, the proportion of participants with abnormal 2HPP increases with age, the highest seen among the middle-age group 45 to 54 (54.5%). Also, the proportion of participants with dysglycemia with a low sputum load was 19.4% compared to 20% in the case of normoglycemic patients whereas 80.6% and 79.8% of participants with high sputum load were dysglycemic and normoglycemic, respectively. These marginal differences did not show any significant differences. Full details of the demographic characteristics are shown in Table 1.

Table 1.
Sociodemographic, lifestyle, and clinical characteristics associated with 2HPP among TB patients

DEMOGRAPHIC CHARACTERISTICS 2HPP CATEGORY TOTAL χ²

ABNORMAL NORMAL

N (%) N (%) N

Age group (years) 171 10 . 29 *

18 to 24 8 (22.2) 28 (77.8) 36

25 to 34 13 (31.0) 29 (69.0) 42

35 to 44 15 (48.4) 16 (51.6) 31

45 to 54 18 (54.5) 15 (45.5) 33

55 to 64 13 (44.8) 16 (55.2) 29

Sex 171 0 . 65

Male 55 (40.7) 80 (59.3) 135

Female 12 (33.3) 24 (66.7) 36

Marital status 171 1 . 24

Married 30 (40.0) 45(60.0) 75

Single 32 (36.8) 55 (63.2) 87

Separated/divorced 5 (55.6) 4 (44.4) 9

Educational level 171 9 . 65 *

None 5 (33.3) 10 (66.7) 15

Primary 6 (37.5) 10 (62.5) 16

Middle/JHS 34 (54.0) 29 (46.0) 63

SHS/O level 14 (28.6) 35 (71.4) 49

Tertiary 8 (28.6) 20 (71.4) 28

Employment status 170 1 . 79

No 13 (30.2) 30 (69.8) 43

Yes 53 (41.7) 74 (58.3) 127

Smoking status 168 0 . 08

Yes 5 (35.7) 9 (64.3) 14

No 61 (39.6) 93 (60.4) 154

Alcohol intake 167 0 . 77

Yes 9 (32.1) 19 (67.9) 28

No 57 (41.0) 82 (59.0) 139

BMI 157 1 . 78

Below 18.5 (underweight) 45(42.1) 62 (57.9) 107

18.5 to 24.9(normal) 15 (33.3) 30 (66.7) 45

25 to 29.9 (overweight) 1 (20.0) 4 (80.0) 5

Systolic BP 170 0 . 88

Normal 43 (39.8) 65 (60.2) 108

Elevated 16 (34.0) 31 (66.0) 47

Hypertension 7 (46.7) 8 (53.3) 15

Diastolic BP 170 0 . 21

Normal 59 (39.3) 91 (60.7) 150

Elevated 5 (33.3) 10 (66.7) 15

Hypertension 2 (40.0) 3 (60.0) 5

Waist–hip ratio 171 0 . 28

Low 55 (40.1) 82 (59.9) 137

Moderate 5 (35.7) 9 (64.3) 14

High 7 (35.0) 13 (65.0) 20

Sputum load (2HPP) 171 0 . 016

Low Load 13 (19.4) 21 (20.2) 34

High Load 54 (80.6) 83 (79.8) 137

DEMOGRAPHIC CHARACTERISTICS	2HPP CATEGORY	TOTAL	χ²
Age group (years)			171	10 . 29 *
18 to 24	8 (22.2)	28 (77.8)	36
25 to 34	13 (31.0)	29 (69.0)	42
35 to 44	15 (48.4)	16 (51.6)	31
45 to 54	18 (54.5)	15 (45.5)	33
55 to 64	13 (44.8)	16 (55.2)	29
Sex			171	0 . 65
Male	55 (40.7)	80 (59.3)	135
Female	12 (33.3)	24 (66.7)	36
Marital status			171	1 . 24
Married	30 (40.0)	45(60.0)	75
Single	32 (36.8)	55 (63.2)	87
Separated/divorced	5 (55.6)	4 (44.4)	9
Educational level			171	9 . 65 *
None	5 (33.3)	10 (66.7)	15
Primary	6 (37.5)	10 (62.5)	16
Middle/JHS	34 (54.0)	29 (46.0)	63
SHS/O level	14 (28.6)	35 (71.4)	49
Tertiary	8 (28.6)	20 (71.4)	28
Employment status			170	1 . 79
No	13 (30.2)	30 (69.8)	43
Yes	53 (41.7)	74 (58.3)	127
Smoking status			168	0 . 08
Yes	5 (35.7)	9 (64.3)	14
No	61 (39.6)	93 (60.4)	154
Alcohol intake			167	0 . 77
Yes	9 (32.1)	19 (67.9)	28
No	57 (41.0)	82 (59.0)	139
BMI			157	1 . 78
Below 18.5 (underweight)	45(42.1)	62 (57.9)	107
18.5 to 24.9(normal)	15 (33.3)	30 (66.7)	45
25 to 29.9 (overweight)	1 (20.0)	4 (80.0)	5
Systolic BP			170	0 . 88
Normal	43 (39.8)	65 (60.2)	108
Elevated	16 (34.0)	31 (66.0)	47
Hypertension	7 (46.7)	8 (53.3)	15
Diastolic BP			170	0 . 21
Normal	59 (39.3)	91 (60.7)	150
Elevated	5 (33.3)	10 (66.7)	15
Hypertension	2 (40.0)	3 (60.0)	5
Waist–hip ratio			171	0 . 28
Low	55 (40.1)	82 (59.9)	137
Moderate	5 (35.7)	9 (64.3)	14
High	7 (35.0)	13 (65.0)	20
Sputum load (2HPP)			171	0 . 016
Low Load	13 (19.4)	21 (20.2)	34
High Load	54 (80.6)	83 (79.8)	137

*P < .05.

Abbreviations: 2HPP, 2-hour post-prandial glucose; TB, tuberculosis; BP, blood pressure; BMI, body mass index; SHS, senior high school;

3.2 Glycemic Variables of Participants

Of a total of 171 participants, using FPG values 87.1%, 8.8%, and 4.1% had normal glucose, impaired fasting, and diabetes, respectively. Using 2HPP values, 61.4%, 27.5%, and 11.1% had normal glucose, impaired glucose tolerance, and diabetes, respectively. The combined prevalence of dysglycemia was thus 12.9% (n = 22) and 38.6% (n = 66) using fasting glucose and 2HPP values, respectively, as shown in Table 2.

Table 2.
Glycaemic variables of study participants

VARIABLE FREQUENCY (N) PROPORTION (%)

Fasting Plasma Glucose (mmol/L)

(Mean ± SD) 5.21 ± 1.46

Normal (<6.1) 149 87.1

Impaired/borderline (6.1–7) 15 8.8

Diabetes (>7.1) 7 4.1

2 HPP Glucose (mmol/L)

(Mean ± SD) 8.24 ± 3.26

Normal (<6.1) 105 61.4

Impaired/borderline (6.1–7) 47 27.5

Diabetes (>7.1) 19 11.1

VARIABLE	FREQUENCY (N)	PROPORTION (%)
Fasting Plasma Glucose (mmol/L)
(Mean ± SD)	5.21 ± 1.46
Normal (<6.1)	149	87.1
Impaired/borderline (6.1–7)	15	8.8
Diabetes (>7.1)	7	4.1
2 HPP Glucose (mmol/L)
(Mean ± SD)	8.24 ± 3.26
Normal (<6.1)	105	61.4
Impaired/borderline (6.1–7)	47	27.5
Diabetes (>7.1)	19	11.1

Abbreviation: 2-HPP, 2-hour post-prandial glucose.

3.3 Predictors of Abnormal Glucose Patterns Among Study Participants

Factors influencing the development of abnormal glucose were explored using binary logistic regression. Being aged 45 to 54 years had 6.43 increased odds (P-value = .007) for developing abnormal glucose compared to the reference age of 18 to 24 years. There were no other significant predictors of abnormal glucose values as shown in Table 3.

Table 3.
Factors influencing abnormal glycaemia among TB patients

COVARIATES AOR P-VALUE 95% CONF. INTERVAL

LOWER UPPER

Age group

18 to 24 Ref

25 to 34 1.66 .398 0.51 5.38

35 to 44 3.27 .076 0.88 12.12

45 to 54 6.43 .007 * 1.66 24.86

55 to 64 2.90 .144 0.70 12.11

Sex

Male Ref

Female 0.80 .767 0.18 3.50

Marital status

Married Ref

Single 1.34 .504 0.57 3.19

Separated/divorced 1.10 .900 0.24 5.07

Educational level

None Ref

Primary 1.66 .586 0.27 10.26

Middle/Junior High School 3.25 .125 0.72 14.68

Senior High School/O-level 1.08 .926 0.22 5.18

Tertiary 1.02 .977 0.20 5.30

Employment status

No Ref

Yes 2.26 .052 0.99 5.14

Smoking status

Yes Ref

No 0.59 .460 0.14 2.40

Alcohol intake

Yes Ref

No 0.56 .299 0.18 1.68

Systolic BP

Normal Ref

Elevated 1.05 .901 0.47 2.37

Hypertension 1.41 .663 0.30 6.61

Diastolic BP

Normal Ref

Elevated 0.42 .250 0.09 1.86

Hypertension 0.49 .517 0.06 4.26

Waist-to-hip ratio

Low Ref

Moderate 0.98 .978 0.18 5.17

High 0.90 .900 0.19 4.39

COVARIATES	AOR	P-VALUE	95% CONF. INTERVAL
Age group
18 to 24	Ref
25 to 34	1.66	.398	0.51	5.38
35 to 44	3.27	.076	0.88	12.12
45 to 54	6.43	.007 *	1.66	24.86
55 to 64	2.90	.144	0.70	12.11
Sex
Male	Ref
Female	0.80	.767	0.18	3.50
Marital status
Married	Ref
Single	1.34	.504	0.57	3.19
Separated/divorced	1.10	.900	0.24	5.07
Educational level
None	Ref
Primary	1.66	.586	0.27	10.26
Middle/Junior High School	3.25	.125	0.72	14.68
Senior High School/O-level	1.08	.926	0.22	5.18
Tertiary	1.02	.977	0.20	5.30
Employment status
No	Ref
Yes	2.26	.052	0.99	5.14
Smoking status
Yes	Ref
No	0.59	.460	0.14	2.40
Alcohol intake
Yes	Ref
No	0.56	.299	0.18	1.68
Systolic BP
Normal	Ref
Elevated	1.05	.901	0.47	2.37
Hypertension	1.41	.663	0.30	6.61
Diastolic BP
Normal	Ref
Elevated	0.42	.250	0.09	1.86
Hypertension	0.49	.517	0.06	4.26
Waist-to-hip ratio
Low	Ref
Moderate	0.98	.978	0.18	5.17
High	0.90	.900	0.19	4.39

P < .05.

Abbreviations: aOR, adjusted odds ratio; TB, tuberculosis; SHS/O level, senior high school/ordinary level.

3.4 Comparison of Sputum Conversion Rates at 2 and 6 Months

The sputum status was assessed and compared to determine the sputum conversion rates at 2 and 6 months. At 2 months, a significantly higher proportion of normoglycemic subjects had negative sputum compared with those with dysglycemia (83% vs 67%) (Z-test = −2.5; P < .05). Despite the proportionally higher numbers of the dysglycemia group that maintained sputum conversion at 6 months, there was no statistical difference between the 2 groups (87% vs 77%, Z-test = −1.71, P > .05), as shown in Table 4.

Table 4.
Test of proportion showing the differences in negative sputum test result between TB patients subjects with abnormal glucose and those with normal glucose (using 2HPP) at two and six months (n = 171)

2 HPP 2 MONTHS 6 MONTHS

PROPORTION (95% CI) PROPORTION (95% CI)

Abnormal 67.16 (55.92–78.41) 77.61 (67.63–87.59)

Normal 83.65 (76.55–90.76) 87.5 (81.14–93.85)

Z-statistic −2.51 * −1.71

2 HPP	2 MONTHS	6 MONTHS
Abnormal	67.16 (55.92–78.41)	77.61 (67.63–87.59)
Normal	83.65 (76.55–90.76)	87.5 (81.14–93.85)
Z-statistic	−2.51 *	−1.71

P-value < .05.

Abbreviations: 2HPP, 2-hour post-prandial glucose; TB, tuberculosis

3.5 Predictors of Sputum Conversion at 2 and 6 Months

Binary logistic regression models were used to test the effect of sociodemographic, clinical, and lifestyle factors on the treatment (sputum conversion) (Table 5). After controlling for demographic factors and adjusting for age groups, the following factors were associated with sputum conversion at 2 months: subjects with dysglycemia were 66% less likely than those with normoglycemia (OR (adjusted odds ratio (aOR)), [95% confidence interval (CI)], P-value: 0.34, [0.14–0.82], .018); females were 7.44 times more likely than men (aOR, [95% CI], P-value: 7.44, [1.73–32.0], .007); and those with high WHR were 88% less likely compared with those with low WHR (aOR, [95% CI], P-value: 0.12, [0.02–0.54], .006) to convert sputum, respectively. At 6 months and after similar adjustments, females (compared with males) and high WHR (compared with those with normal WHR) were 9.29 times more likely (aOR, [95% CI], P-value: 7.44, 9.29 [1.78–48.5], .008] and 89% less likely (aOR, [95% CI], P-value: 0.11, [0.02–0.63], .013) for sputum conversion, respectively.

Table 5.
2HPP influence on treatment outcome (sputum conversion) among TB patients adjusting for demographic factors and controlling for age group

Covariates PERIOD OF TREATMENT OUTCOME

2 MONTHS 6 MONTHS

AOR (95% CI) P-VALUE AOR (95% CI) P-VALUE

Two HPP

Normal Ref Ref

Abnormal 0.34 (0.14–0.82) .018 0.46 (0.17–1.25) .128

Sex

Male Ref Ref

Female 7.44 (1.73–32.0) .007 9.29 (1.78–48.5) .008

Marital status

Married Ref Ref

Single 1.47 (0.64–to 3.41) .363 1.75 (0.06–4.52) .249

Separated/divorced 0.41 (0.08–1.91) .254 0.42 (0.06–2.81) .369

Educational level

None Ref Ref

Primary 0.44 (0.08–2.49) .358 0.74 (0.08–6.46] .785

Middle/Junior High School 1.19 [(0.25–5.36) .826 3.82 (0.57–25.52) .166

Senior High School/O-level 0.84 (0.18–3.82) .819 2.00 (0.33–12.14) .453

Tertiary 0.72 (0.15–3.37) .679 0.72 (0.13–4.04) .713

Employment status

No Ref Ref

Yes 1.27 (0.50–3.21) .619 0.22 (0.04–1.14) .072

Smoking status

Yes Ref Ref

No 0.49 (0.08–2.91) .435 1.94 (0.31–12.27) .482

Alcohol intake

Yes Ref Ref

No 1.67 (0.50–5.58) .401 2.21 (0.26–5.57) .809

Waist-to-hip ratio

Low Ref Ref

Moderate 0.48 (0.09–2.44) .374 0.81 (0.11–5.95) .837

High 0.12 (0.02–0054) .006 0.11 (0.02–0.63) .013

Covariates	PERIOD OF TREATMENT OUTCOME
Two HPP
Normal	Ref	Ref
Abnormal	0.34 (0.14–0.82) .018	0.46 (0.17–1.25) .128
Sex
Male	Ref	Ref
Female	7.44 (1.73–32.0) .007	9.29 (1.78–48.5) .008
Marital status
Married	Ref	Ref
Single	1.47 (0.64–to 3.41) .363	1.75 (0.06–4.52) .249
Separated/divorced	0.41 (0.08–1.91) .254	0.42 (0.06–2.81) .369
Educational level
None	Ref	Ref
Primary	0.44 (0.08–2.49) .358	0.74 (0.08–6.46] .785
Middle/Junior High School	1.19 [(0.25–5.36) .826	3.82 (0.57–25.52) .166
Senior High School/O-level	0.84 (0.18–3.82) .819	2.00 (0.33–12.14) .453
Tertiary	0.72 (0.15–3.37) .679	0.72 (0.13–4.04) .713
Employment status
No	Ref	Ref
Yes	1.27 (0.50–3.21) .619	0.22 (0.04–1.14) .072
Smoking status
Yes	Ref	Ref
No	0.49 (0.08–2.91) .435	1.94 (0.31–12.27) .482
Alcohol intake
Yes	Ref	Ref
No	1.67 (0.50–5.58) .401	2.21 (0.26–5.57) .809
Waist-to-hip ratio
Low	Ref	Ref
Moderate	0.48 (0.09–2.44) .374	0.81 (0.11–5.95) .837
High	0.12 (0.02–0054) .006	0.11 (0.02–0.63) .013

Abbreviations: 2HPP, 2-hour post-prandial glucose; aOR, adjusted odds ratio; SHS, senior high school TB, tuberculosis

4. Discussion

The results of the study show that at 2 months, a significantly higher proportion of normoglycemic subjects had negative sputum compared with those with dysglycemia. However, despite the proportionally higher numbers of the normoglycemia group that maintained sputum conversion at 6 months, there was no statistical difference between the 2 groups. This suggests a transient impact of dysglycemia on sputum conversion among TB patients. While some studies have supported the negative impact of diabetes on sputum conversion throughout the period of TB treatment, others have showed a transient impact.^9,11,12 Alisjahbana et al. reported that TB patients who were diabetic had significantly higher percent of smear-positive results than their nondiabetic counterparts at 2 months (18.1% vs 10.0%). Unlike our study, after 6 months (end of the treatment), 22.2% of sputum specimens from diabetic patients were positive for M tuberculosis ( aOR, 7.65; P = .004).⁹ In a retrospective study that assessed TB patients with self-reported diabetes from south Texas (USA) and north-eastern Mexico, the authors reported that TB patients were more likely to remain positive at the end of the first (Texas cohort) or second (Mexico cohort) month of treatment,¹² which was similar to our study. A systematic review published in 2011 by Baker et al. reported an increased risk of relapse in these patients (relative risk, 3.89; 95% CI, 2.43–6.23). Unlike our study, some studies have however shown no relationship between diabetes and sputum conversion rate at the end of month 2,^13,16,17 while a trend toward increased time to sputum conversion has rather been found in other studies.^15,26,27

The transient impact of dysglycemia on sputum conversion rate suggests possible nonlasting pathophysiological linkages and putative mechanisms.^28,29 The risk of TB patients with diabetes for reduced sputum conversion rate is suggested putatively due to the reduced numbers as well as the function of T lymphocytes involved in T-helper cells 1(TH1) cytokine inhibition of M tuberculosis.^30,31 In diabetes, dysfunction of macrophages occurs, which impairs phagocytic and chemotactic function as well as the production of reactive oxygen species.^30,31 Further, the impairment of chemotaxis of monocytes is thought to occur in diabetes.³² The respiratory burst used in expelling pathogens is also thought to be impaired with diabetes.^30,31

TB on its own has been linked to the development of impaired glucose tolerance (IGT)^28,33 and new-onset diabetes.^28,34 While IGT generally normalizes after TB has been treated, the risk of developing type 2 diabetes in the future remains high.³⁵ Some other studies revealed that between 19% to 42.6% of active TB patients, who were discovered to have IGT or diabetes, had significant reduction or complete regression in the rates after treatment.^28,29 This suggested stress response to infection leading to dysglycemia is thought to be due to the elaboration of interleukin 1 (IL-1), interleukin 6 (IL-6), and tumor necrosis factor-alpha.^5,35

Unsurprisingly in our study, most participants were young, male, and single; those are well-known sociodemographic characteristics of TB patients.^1,36,37 Other risk factors for TB include low incomes, the load of the bacilli, malnutrition, excessive alcohol use, overcrowding, smoking, HIV infection, diabetes, drugs that cause immunosuppression, and closeness to a person with active TB.^1,5–36

The increased risk of the age group 45–54 years with abnormal blood glucose (aOR, 6.43) may represent an increased inherent risk for dysglycemia and type 2 diabetes conferred by increasing age.³⁸ Subjects with high WHR were less likely to convert sputum compared with those with normal WHR both at 2 and 6 months. High WHR is a risk factor for dysglemia and diabetes,³⁹ which influences the development of reduced clearance of TB pathogens as described earlier on.^30,31 This association is also independent of age, family history of diabetes, and sex;³⁹ the sex association is stronger in women than men.³⁹ Central obesity as measured by parameters such as WHR, waist circumference (WC), and BMI has been associated with impaired glucose–insulin homeostasis and insulin clearance, decreased insulin-stimulated glucose disposal finally which leads to decreased glucose tolerance.⁴⁰

It is not clear why females were more likely to convert sputum than male counterparts both at 2 and 6 months. It must, however, be stated that females are more likely than men to be compliant with their treatment and also keep their appointments, which may in turn improve treatment outcomes.^41,42

5. Conclusion

This study has added to the body of knowledge on the impact of dysglycemia on sputum conversion among TB patients. A significantly lower proportion of smear-positive TB patients with dysglycemia converted to smear negative after 2 months of treatment but not at the end of treatment suggesting a transient impact of dysglycemia on sputum conversion. Other factors, including age, female sex, and increased WHR, have shown association. Larger studies are needed to validate these findings. A change in the current TB regimen is not recommended based on our findings.

Future research is required in this field and must compare the severity of TB presentation between the dysglycemia and normoglycemic groups, such as symptom burden, pattern of lobe involvement, reactivation of old foci, rates of hemoptysis, fever, and atypical presentations and cavitations. We plan to perform formal glucose tolerance tests at 2 months (end of the intensive phase) and at 6 months (end of the treatment) to confirm whether dysglycemia associated with TB is transient. Patients would also be followed up to ascertain the RRs between the two groups.

Footnotes

Acknowledgments

We also appreciate the contribution of Ama Aidoo, Norah Nkornu, and Kelvin Acquaye, especially in data collection. Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon request

Author Contributions

EY conceived the study, participated in its design, data collection, analysis, and drafted the manuscript and collation of all drafts. VB, IDD, VG, MBAM, EKA, JT, and CCM contributed to the study design, data collection, analysis, and writing the manuscript draft. All authors read and approved the final version of the manuscript.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical Approval and Consent to Participate

Ethical and Protocol approval for the study was sought from the College of Health Sciences Ethics and Protocol Review Committee of the University of Ghana with reference number URF/9/ILG-076/2015 to 2016. It complied with the Helsinki Declaration of 1964 (Revised 2013) on human experimentation. All patients provided written informed consent. Strict confidentiality of data and privacy for study participants were ensured. Data were kept secured and available only to the principal investigator.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the University of Ghana Office of Research, Innovation and Development (ORID) grant (Grant no. URF/9/ILG-076/2015-2016).

ORCID iD

Ernest Yorke

Trial Registration

Not applicable, as no trial was conducted.

Ernest Yorke, FWACP, FGCP, MSc, Diabetes (Cardiff), MSc, Endo. (UK), MB ChB, BSc, Cert. HAM (Ghana Institute of Management and Public Administration, GIMPA). Dr Ernest Yorke is a fellow of the West African College of Physicians and the Ghana College of Physicians & Surgeons. Currently, he is a senior lecturer at the University of Ghana Medicine School, and a consultant physician/endocrinologist at the Korle-Bu Teaching Hospital, Accra. He also holds double Masters' degrees from universities from the United Kingdom. He has been extensively involved in the organization and also served as a facilitator to numerous metabolic and cardiovascular Continuing Medical Education programs, both locally and abroad. He values the use of practical tools as well as new technology in research and development to solve problems and also to advance knowledge. He has served as a consultant and Advisory Board member for a number of multinational pharmaceutical companies involved in diabetes care, both locally and abroad. His main research interests include diabetes and kidney disease, psychological impact of diabetes, pregnancy outcomes in diabetes, and diabetic foot care as well as thyroid diseases. He has over 30 peer-reviewed publications to his credit. He has other interests outside clinical work and academia. Among many other volunteer works, he is currently the Chairman of the Greater Accra Division of the Ghana Medical Association and also a Board member of Diabetes Youth Care, a non-governmental organization dedicated to the care and well-being of young people living with diabetes in Ghana. He is married with three children and watches TV in his spare time.

References

WHO. Global Tuberculosis Report 2017. World Health Organization; 2017.

Young

Wotton

Critchley

Unwin

Goldacre

. Increased risk of tuberculosis disease in people with diabetes mellitus: record-linkage study in a UK population. J Epidemiol Community Health. 2010;66(6):519-523. doi:10.1136/jech.2010.114595

Yorke

Atiase

Akpalu

Sarfo-Kantanka

Boima

Dey

. The bidirectional relationship between Tuberculosis and diabetes. Tuberc Res Treat. 2017;6:1702578. doi:10.1155/2017/1702578

Saeedi P,

Petersohn I

, Salpea P, et al. Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045: Results from the International Diabetes Federation Diabetes Atlas, 9th edition. DRCP. 2019;157(107843):1-10.

Niazi

Kalra

. Diabetes and tuberculosis: a review of the role of optimal glycemic control. J Diabetes Metab Disord. 2012;11(1):2251–6581.

Stevenson

Critchley

Forouhi

, et al.

Diabetes and the risk of tuberculosis: a neglected threat to public health?

Chronic Illn. 2007;3(3):228–245.

Pealing

Wing

Mathur

Prieto-Merino

Smeeth

Moore

DAJ

. Risk of tuberculosis in patients with diabetes: population based cohort study using the UK clinical practice research datalink. BMC Med. 2015;13(1):1–16.

Noubiap

Nansseu

Nyaga

, et al. Global prevalence of diabetes in active tuberculosis: a systematic review and meta-analysis of data from 2·to 3 million patients with tuberculosis. Lancet Glob Health. 2019;7(4):e448–e460.

Alisjahbana

Sahiratmadja

Nelwan

, et al. The effect of type 2 diabetes mellitus on the presentation and treatment response of pulmonary tuberculosis. Clin Infect Dis. 2007;45(4):428–435.

10.

Wilson

. Infection and diabetes mellitus. In: Pickup

Williams

, eds. Textbook of Diabetes. Blackwell Scientific Publication; 1991:813–819.

11.

Baker

Harries

Jeon

, et al. The impact of diabetes on tuberculosis treatment outcomes: a systematic review. BMC Med. 2011;9(81):1741–7015.

12.

Restrepo

Fisher-Hoch

Crespo

, et al. Type 2 diabetes and tuberculosis in a dynamic bi-national border population. Epidemiol Infect. 2007;135(3):483–491.

13.

Dooley

Tang

Golub

Dorman

Cronin

. Impact of diabetes mellitus on treatment outcomes of patients with active tuberculosis. Am J Trop Med Hyg. 2009;80(4):634–639.

14.

Dooley

Chaisson

. Tuberculosis and diabetes mellitus: convergence of two epidemics. Lancet Infect Dis. 2009;9(12):737–746.

15.

Gautam

Shrestha

Mahato

Nguyen

Mishra

Berg-Beckhoff

. Diabetes among tuberculosis patients and its impact on tuberculosis treatment in South Asia: a systematic review and meta-analysis. Sci Rep. 2021;11(1):1–12.

16.

Kameda

Kawabata

Masuda

. Follow-up study of short course chemotherapy for pulmonary tuberculosis complicated with diabetes mellitus. Kekkaku (Tuberculosis). 1990;65(12):791–803.

17.

Singla

Khan

Al-Sharif

Al-Sayegh

Shaikh

Osman

. Influence of diabetes on manifestations and treatment outcome of pulmonary TB patients. Int J Tuberc Lung Dis. 2006;10(1):74–79.

18.

Cheng

Zhang

Zhao

Wang

Chen

. Mutual impact of diabetes mellitus and tuberculosis in China. Biomed Environ Sci. 2017;30(5):384–389.

19.

Chiang

Bai

Lin

, et al. The influence of diabetes, glycemic control, and diabetes-related comorbidities on pulmonary tuberculosis. PloS one. 2015;10(3):e0121698

20.

Morsy

Zaher

Hassan

Shouman

. Predictors of treatment failure among tuberculosis patients under DOTS strategy in Egypt. East Mediterr Health J. 2003 Jul;9(4):689–701.

21.

Jiménez-Corona

Cruz-Hervert

García-García

, et al. Association of diabetes and tuberculosis: impact on treatment and post-treatment outcomes. Thorax. 2013;68(3):214–220.

22.

Singla

Osman

Khan

Al-Sharif

Al-Sayegh

Shaikh

. Factors predicting persistent sputum smear positivity among pulmonary tuberculosis patients 2 months after treatment. Int J Tuberc Lung Dis. 2003;7(1):58–64.

23.

Sacks

Bruns

Goldstein

Maclaren

McDonald

Parrott

. Guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus. Clin Chem. 2002;48(3):436–472.

24.

World Health Organization. International Diabetes Federation (2006) Definition and diagnosis of diabetes mellitus and intermediate hyperglycemia: report of a WHO/IDF consultation. IDF Consultation; 2008.

25.

Alberti

Zimmet

Shaw

Grundy

. The IDF consensus worldwide definition of the metabolic syndrome. Brussels: Int Diabetes Fed. 2006;23(5):469–480.

26.

Heysell

Moore

Keller

Houpt

. Therapeutic drug monitoring for slow response to tuberculosis treatment in a state control program, Virginia, USA. Emerg Infect Dis. 2010;16(10):1546.

27.

Chang

J-T

Dou

H-Y

Yen

C-L

, et al. Effect of type 2 diabetes mellitus on the clinical severity and treatment outcome in patients with pulmonary tuberculosis: a potential role in the emergence of multidrug-resistance. J Formos Med Assoc. 2011;110(6):372–381.

28.

Basoglu

Bacakoglu

Cok

Saymer

Ates

. The oral glucose tolerance test in patients with respiratory infections. Monaldi Arch Chest Dis. 1999;54(4):307–310.

29.

Oluboyo

Erasmus

. The significance of glucose intolerance in pulmonary tuberculosis. Tubercle. 1990;71(2):135–138.

30.

Geerlings

Hoepelman

. Immune dysfunction in patients with diabetes mellitus (DM). FEMS Immunol Med Microbiol. 1999;26(3-4):259–265.

31.

Tsukaguchi

Yoneda

Yoshikawa

, et al. Case study of interleukin-1 beta, tumor necrosis factor alpha and interleukin-6 production peripheral blood monocytes in patients with diabetes mellitus complicated by pulmonary tuberculosis. Kekkaku. 1992;67(12):755–760.

32.

Moutschen

Scheen

Lefebvre

. Impaired immune responses in diabetes mellitus: analysis of the factors and mechanisms involved. Relevance to the increased susceptibility of diabetic patients to specific infections. Diabete Metab. 1992;18(3):187–201.

33.

Jeon

Harries

Baker

, et al. Bi-directional screening for tuberculosis and diabetes: a systematic review. Research support, Non-US Govt review. Trop Med Int Health. 2010;15(11):1300–1314.

34.

Mugusi

Swai

Alberti

McLarty

. Increased prevalence of diabetes mellitus in patients with pulmonary tuberculosis in Tanzania. Tubercle. 1990;71(4):271–276.

35.

Amrit

Ashok

. Tuberculosis and diabetes: an appraisal. Indian J Tuberc. 2000;47(1):3–8.

36.

Amare

Gelaw

Anagaw

Gelaw

. Smear positive pulmonary tuberculosis among diabetic patients at the Dessie referral hospital, northeast Ethiopia. Infect Dis Poverty. 2013;2(1):6.

37.

Narasimhan

Wood

MacIntyre

Mathai

. Risk factors for tuberculosis. Pulm Med. 2013;2013. doi: 10.1155/2013/828939

38.

Kirkman

Briscoe

Clark

, et al. Diabetes in older adults. Diabetes Care. 2012;35(12):2650–2664. doi:10.2337/dc12-1801

39.

Schmidt

Duncan

Canani

Karohl

Chambless

. Association of waist-hip ratio with diabetes mellitus: strength and possible modifiers. Diabetes Care. 1992;15(7):912–914.

40.

Vazquez

Duval

Jacobs Jr

Silventoinen

. Comparison of body mass index, waist circumference, and waist/hip ratio in predicting incident diabetes: a meta-analysis. Epidemiol Rev. 2007;29(1):115–128.

41.

Manteuffel

Williams

Chen

Verbrugge

Pittman

Steinkellner

. Influence of patient sex and gender on medication use, adherence, and prescribing alignment with guidelines. J Womens Health. Feb 2014;23(2):112–119. doi:10.1089/jwh.2012.3972

42.

Rodriguez Pacheco

Negro Alvarez

Campuzano Lopez

, et al. Non-compliance with appointments amongst patients attending an allergology clinic, after implementation of an improvement plan. Allergol Immunopathol (Madr). Jul-Aug 2007;35(4):136–144.

COVARIATES	AOR	P-VALUE	95% CONF. INTERVAL
COVARIATES	AOR	P-VALUE	LOWER	UPPER
Age group
18 to 24	Ref
25 to 34	1.66	.398	0.51	5.38
35 to 44	3.27	.076	0.88	12.12
45 to 54	6.43	.007 *	1.66	24.86
55 to 64	2.90	.144	0.70	12.11
Sex
Male	Ref
Female	0.80	.767	0.18	3.50
Marital status
Married	Ref
Single	1.34	.504	0.57	3.19
Separated/divorced	1.10	.900	0.24	5.07
Educational level
None	Ref
Primary	1.66	.586	0.27	10.26
Middle/Junior High School	3.25	.125	0.72	14.68
Senior High School/O-level	1.08	.926	0.22	5.18
Tertiary	1.02	.977	0.20	5.30
Employment status
No	Ref
Yes	2.26	.052	0.99	5.14
Smoking status
Yes	Ref
No	0.59	.460	0.14	2.40
Alcohol intake
Yes	Ref
No	0.56	.299	0.18	1.68
Systolic BP
Normal	Ref
Elevated	1.05	.901	0.47	2.37
Hypertension	1.41	.663	0.30	6.61
Diastolic BP
Normal	Ref
Elevated	0.42	.250	0.09	1.86
Hypertension	0.49	.517	0.06	4.26
Waist-to-hip ratio
Low	Ref
Moderate	0.98	.978	0.18	5.17
High	0.90	.900	0.19	4.39

2 HPP	2 MONTHS	6 MONTHS
2 HPP	PROPORTION (95% CI)	PROPORTION (95% CI)
Abnormal	67.16 (55.92–78.41)	77.61 (67.63–87.59)
Normal	83.65 (76.55–90.76)	87.5 (81.14–93.85)
Z-statistic	−2.51 *	−1.71

Covariates	PERIOD OF TREATMENT OUTCOME
	2 MONTHS	6 MONTHS
	AOR (95% CI) P-VALUE	AOR (95% CI) P-VALUE
Two HPP
Normal	Ref	Ref
Abnormal	0.34 (0.14–0.82) .018	0.46 (0.17–1.25) .128
Sex
Male	Ref	Ref
Female	7.44 (1.73–32.0) .007	9.29 (1.78–48.5) .008
Marital status
Married	Ref	Ref
Single	1.47 (0.64–to 3.41) .363	1.75 (0.06–4.52) .249
Separated/divorced	0.41 (0.08–1.91) .254	0.42 (0.06–2.81) .369
Educational level
None	Ref	Ref
Primary	0.44 (0.08–2.49) .358	0.74 (0.08–6.46] .785
Middle/Junior High School	1.19 [(0.25–5.36) .826	3.82 (0.57–25.52) .166
Senior High School/O-level	0.84 (0.18–3.82) .819	2.00 (0.33–12.14) .453
Tertiary	0.72 (0.15–3.37) .679	0.72 (0.13–4.04) .713
Employment status
No	Ref	Ref
Yes	1.27 (0.50–3.21) .619	0.22 (0.04–1.14) .072
Smoking status
Yes	Ref	Ref
No	0.49 (0.08–2.91) .435	1.94 (0.31–12.27) .482
Alcohol intake
Yes	Ref	Ref
No	1.67 (0.50–5.58) .401	2.21 (0.26–5.57) .809
Waist-to-hip ratio
Low	Ref	Ref
Moderate	0.48 (0.09–2.44) .374	0.81 (0.11–5.95) .837
High	0.12 (0.02–0054) .006	0.11 (0.02–0.63) .013