Table 1 Systematic review of 9 studies reporting prevalence of CKD in patients with COPD compared to controls; included in meta-analysis

Baty et al.; 2013 [8]

Study design: Population based case-control study

Funding: Takeda Pharma AG, Switzerland

Setting & period: All hospitalizations in Switzerland between 2002 and 2010

COPD group: 340, 948 patients, 64% males, median age 73 years

Non-COPD group: 340,948 patients, 64% males, median age 73 years

Diagnosis of COPD: Based on ICD-10 codes

CKD definition: Based on ICD-10 code

Blinding of outcome adjudicator: not reported

Selection bias: none

Information bias: objective

outcome evaluation: no;

standardized CKD risk

measurement: no

Confounding: no Matching: yes. Adjustment in analysis: yes Confounding variables: no

Loss to follow up: none

4.39% of patients with COPD had Chronic kidney disease (ICD 10 code, N188) compared to 2.13% of patients without COPD (p < 0.001)

4.64% of patients with COPD had Chronic kidney disease unspecified (ICD 10 code, N189) compared to 2.25% of patients without COPD (p < 0.001)

Gjerde et al.; 2011 [9]

Study design: Case-control study

Funding: The Foundation for Respiratory Research, Center for Clinical Research, Bergen

Setting & period: Patients aged 40-76 years with COPD were recruited from health institutions in Hordaland County in Western Norway, where as those without COPD were recruited among former participants from a general population survey in Hordaland County; between 2006 and 2007

COPD group: 433 patients, 59.6% male

Non-COPD group: 233 patients

Diagnosis of COPD: using Spirometry

CKD definition: eGFR <60

Blinding of outcome adjudicator: not reported

Selection bias: yes, voluntarily included, not random

Information bias: objective

outcome evaluation: yes;

standardized CKD risk

measurement: yes

Confounding: yes Matching: no. Adjustment in analysis: yes Confounding variables: no

Loss to follow up: none

Prevalence of undiagnosed renal failure in the COPD patients was 6.9%, significantly higher than among the subjects without COPD (p < 0.001)

Incalzi et al.; 2010 [10]

Study design: Case-control study

Funding: not reported

Setting & period: Participants aged 65 years and older were recruited from pulmonary medicine outpatient facilities in University of Palermo, Italy

COPD group: 356 patients

Non-COPD group: 290 patients

Diagnosis of COPD: Spirometry

CKD definition: eGFR < 60 using MDRD equation

Blinding of outcome adjudicator: not reported

Selection bias: no

Information bias: objective

outcome evaluation: yes;

standardized CKD risk

measurement: yes

Confounding: no Matching: yes (age) Adjustment in analysis: yes Confounding variables: no

Loss to follow up: none

Overall prevalence of Chronic renal failure was 43.0% in COPD group and 23.4% in non-COPD group (p < 0.001)

Logistic regression analysis revealed significant association between COPD and concealed chronic renal failure (OR: 2.19; CI: 1.17-4.12) and overt chronic renal failure (OR: 1.94; CI: 1.01-4.66)

Joo et al.; 2012 [11]

Study design: Cross-sectional Survey

Funding: Grant of Korea Healthcare Technology R&D project

Setting & period: Database of the fourth Korean Health and Nutrition Examination Survey with a nationally representative sample, during 2008. Aged ≥ 40 years

COPD group: 354 patients, 67.2% male, mean age 64.6 years

Non-COPD group: 1823 patients, 36.9% male, mean age 54.4 years

Diagnosis of COPD: Spirometry, FEV1/FVC < 0.7

CKD definition: patients’ awareness of CKD diagnosis was surveyed

Blinding of outcome adjudicator: not reported

Selection bias: no

Information bias: objective

outcome evaluation: yes, for COPD diagnosis only;

standardized CKD risk

measurement: no

Confounding: yes Matching: no. Adjustment in analysis: yes Confounding variables: gender, mean age

Loss to follow up: none

0.6% of patients in COPD group had Chronic renal failure compared to 0.4% in non-COPD group (p = 0.41, not statistically significant)

Mapel et al.; 2013 [12]

Study design: retrospective case-control cohort analysis

Funding: grant from Pfizer Pharmaceuticals Inc.

Setting & period: patients aged 40 years or older seen in 4 hospitals and a network of outpatient clinics of Lovelace Health Systems (LHS) in New Mexico, USA during the study period 2005-2008

COPD group: 2284 patients of LHS aged 40 or more with COPD and have at least 2 outpatient clinic visits or one hospitalization and enrolled with LHS for at least 12 months during the study period; 47.5% men; mean age 70.3 +/- 9.8 yrs

Non-COPD group: 5959 randomly selected patients without a diagnosis of COPD and be of same age and gender

Diagnosis of COPD: ICD-9 diagnosis code of COPD

CKD definition: ICD-9 codes and abnormal renal function tests

Blinding of outcome adjudicator: not reported

Selection bias: none

Information bias: objective

outcome evaluation: yes;

standardized CKD risk

measurement: yes

Confounding: no Matching: yes (age, gender). Adjustment in analysis: yes Confounding variables: no

Loss to follow up: none

Chronic renal failure was more than three times more prevalent among COPD patients (2.89%) than among controls (0.79%) (p < 0.001)

Nagorni-Obradovic; 2014 [13]

Study design: cross sectional study (case-control analysis)

Funding: Ministry of Education and Science of the Republic of Serbia

Setting & period: 10,013 nationally representative sample of adults aged 40 years or older who participated in multipurpose health survey of population of Serbia in 2006

COPD group: 653 patients, 46.6% male, mean age 62.8 years (SD: 12.4)

Non-COPD group: 9.360 patients, 54.4% male, mean age 59.3 years (SD: 12.2)

Diagnosis of COPD: Self-reported history of chronic bronchitis and emphysema

CKD definition: Self-reported history of chronic renal disease

Blinding of outcome adjudicator: N/A

Selection bias: no

Information bias: objective

outcome evaluation: no;

standardized CKD risk

measurement: no

Confounding: no Matching: no

Adjustment in analysis: yes (age, gender, educational level, smoking) Confounding variables: no

Loss to follow up: n/a

20.6% of COPD patients reported having a diagnosis of chronic renal failure compared to 9.3% of non-COPD patients (p < 0.01)

Schnell et al., 2012 [14]

Study design: cross-sectional study (case-control analysis)

Funding: Johns Hopkins, NCRR and NIH

Setting & period: non-institutionalized civilians in the US aged 45 years or more who participated in the National Health and Nutrition Examination Survey (NHANES) from 1998 through 2008

COPD group: 995 patients, 39.9% males, mean age 62.7 years (CI: 61.7-63.8)

Non-COPD group: 14,828 patients, 47% males, mean age 60 years (CI: 59.6-60.3)

Diagnosis of COPD: positive response in NHANES questions to either chronic bronchitis or emphysema with negative response to current asthma

CKD definition: NHANES question with positive response to eGFR < 60 as calculated using MDRD equation

Blinding of outcome adjudicator: N/A

Selection bias: no

Information bias: objective

outcome evaluation: no;

standardized CKD risk

measurement: yes

Confounding: no Matching: no

Adjustment in analysis: yes Confounding variables: none

Loss to follow up: n/a

16.2% of patients with physician diagnosed COPD reported having low eGFR, compared to 10.5% of patients without physician diagnosed COPD (p < 0.0001)

Van Gestel et al.; 2009 [15]

Study design: cohort study

Funding: none

Setting & period: 3358 patients who underwent elective vascular surgery or lower limb arterial reconstruction surgeries between January 1990 to December 2006

COPD group: 1310 patients

Non-COPD group: 2048 patients

Diagnosis of COPD: post bronchodilator pulmonary function test

CKD definition: based on calculated eGFR <60 estimated using MDRD equation

Blinding of outcome adjudicator: not reported

Selection bias: yes, convenience sample

Information bias: objective

outcome evaluation: yes;

standardized CKD risk

measurement: yes

Confounding: yes. Matching: No. Adjustment in analysis: yes Confounding variables: age, gender, type of surgery, current smoking, previous heart failure, hypertension, Diabetes, hyperlipidemia

Loss to follow up: none

COPD was associated with a higher risk of prevalent CKD even after adjustment for confounding variables – OR: 1.22 (1.03 – 1.44) (p = 0.03)

A borderline significant relationship was observed for mild COPD while moderate COPD was independently associated with CKD. No significant association was found between severe COPD and CKD

Yoshizawa et al.; 2015 [6]

Study design: retrospective case-control cohort analysis

Funding: none

Setting & period: outpatient clinic visits of Kanamecho Hospital, Tokyo, Japan for the study period of May 2011 to April 2012

COPD group: 108 stable COPD patients; 83.3% males; mean age 74.3 ± 7.1 year.

Non-COPD group: 73 patients of the same outpatient practice; 49.3% males; mean age 71.8 ± 7.3 years

Diagnosis of COPD: spirometry reading of FEV1/FVC less than 70% after inhalation of a bronchodilator, and severity of obstruction judged according to the Global Initiative for Chronic Obstructive Lung Disease (GOLD) criteria

CKD definition: eGFR less than 60 mL/min/1.73 m2 as per calculation based on serum Creatinine and serum Cystatin levels separately

Blinding of outcome adjudicator: not reported

Selection bias: no

Information bias: objective

outcome evaluation: yes;

standardized CKD risk

measurement: yes

Confounding: yes. Matching: No. Adjustment in analysis: yes Confounding variables: age, gender, BMI, hypertension, Diabetes, hyperlipidemia

Loss to follow up: none

Prevalence of CKD (using Se Cr for calculation of eGFR) was significantly higher in COPD group - OR: 4.91 (1.94 – 12.46) (p = 0.0004)

Prevalence of CKD (using Se Cys for calculation of eGFR) was significantly higher in COPD group - OR: 6.30 (2.99 – 13.26) (p < 0.0001)