Blood Pressure Dipping Status in Chronic Kidney Disease and Association With Adverse Outcomes: A Systematic Review and Meta‐Analysis

Abstract

Background

Nondipping, <10% nighttime blood pressure reduction, is common in chronic kidney disease (CKD). We conducted a systematic review to determine CKD nondipper prevalence, nondipper risk between patients with and without CKD and adverse events associated with nondippers in CKD.

Methods

CKD population studies undergoing an automated blood pressure monitor were searched for on MEDLINE, Embase, and CENTRAL (December 2023), included if they had nondipper information and excluded if they examined only a specific population. Random‐effects analysis was conducted to determine prevalence and relative risk (RR) with 95% CI. Risk of bias was assessed with the Joanna Briggs Institute score for prevalence studies and Newcastle‐Ottawa Scale for observational studies. A sensitivity analysis excluded studied with a high risk of bias. Heterogeneity was examined using I² and SD.

Results

We included 63 articles with 54 515 patients; 33 825 with CKD and 20 690 without CKD. CKD nondipper prevalence was 0.53 (95% CI, 0.52–0.53, SD=13.5%). Compared with patients with CKD, those without CKD had a lower risk of nondipping status (RR, 0.69 [95% CI, 0.63–0.75], SD=14.1%, P<0.001). Similar findings were seen in the sensitivity analysis. CKD, dippers compared with nondippers, was associated with a lower mortality (RR, 0.60 [95% CI, 0.49–0.72], SD=3.2%, P<0.001), cardiovascular events (RR, 0.63 [95% CI, 0.47–0.85], SD=28.3%, P<0.001), and CKD progression (RR, 0.64 [95% CI 0.55–0.75], SD=14.1%, P<0.001).

Conclusions

Nondipping is prevalent in CKD, with CKD itself a risk factor. Nondipping was associated with mortality, cardiovascular event, and CKD progression. Almost all included studies were retrospective cohorts, with a high heterogeneity found.

Nonstandard Abbreviation and Acronym

ABPM

ambulatory blood pressure monitoring

Clinical Perspective.

What Is New?

• This meta‐analysis, pooling all the existing evidence, shows that nondipping status on a 24‐hour ambulatory blood pressure monitor in patients with chronic kidney disease (CKD) is common and can affect up to 56% of patients with CKD.

• CKD is a risk factor for nondipping status, compared with patients with no CKD, with nondipping status in patients with CKD associated with an increased risk of CKD progression, cardiovascular events and mortality.

What Are the Clinical Implications?

• Given the high prevalence and associated risks, a 24‐hour ambulatory blood pressure monitoring should be considered for patients with CKD, with future studies focusing on intervention, to determine if targeting nondipping status improves clinical outcomes in patients with CKD.

Chronic kidney disease (CKD) is defined as abnormalities of kidney structure or function present for a minimum of 3 months, and is usually based on an estimated glomerular filtration rate of <60 mL/min per 1.73 m² or albuminuria (a urine albumin‐to‐creatinine ratio of ≥30 mg/g). CKD represents a major global health burden, affecting >850 million individuals worldwide, ¹ with an estimated 10% to 13% of the global population affected. ² , ³ In patients with CKD, hypertension is highly prevalent, affecting >60% to 90% of individuals. ⁴ Importantly, the prevalence of hypertension increases progressively with CKD severity, from an estimated 23% in individuals with no CKD to 36% in stage I, 48% in stage II, 60% in stage III, and 84% in stages IV to V. ⁵ This bidirectional relationship, wherein hypertension both contributes to and results from kidney disease, underscores the critical importance of blood pressure evaluation in this population.

Normally, blood pressure (BP) exhibits a predictable circadian rhythm, characterized by a physiological nocturnal decline. Using 24‐hour ambulatory BP monitoring (ABPM), normal “dipping” is defined as a 10% to 20% reduction in BP during sleep compared with daytime readings. ⁶ Conversely, a <10% drop in BP during nighttime characterizes nondipping status. ⁶ In the normal population nondipping can occur in around 18% of individuals ⁷ and a nondipping BP status is associated with increased risk of left ventricular hypertrophy, stroke, cardiovascular events, and incident CKD. ⁸ , ⁹

Patients with CKD are known to have abnormalities in circadian BP regulation, ¹⁰ which can manifest as disrupted nocturnal dipping patterns on ABPM. Prevalence of nondipping status in CKD is thought to be significantly higher than that found in the general population, with some studies showing a range between 50% and 80%. ¹¹ , ¹² However, despite the growing awareness of the significance of BP dipping patterns, there is a paucity of information on the prevalence and consequences of nondipping status in patients with CKD.

We set out to conduct a systematic review and meta‐analysis to determine (1) the prevalence of nondippers in patients with CKD, (2) the prevalence nondipping status in patients with CKD compared with patients with no CKD, and (3) adverse outcomes associated with nondipping compared with normal dipping status in patients with CKD, including mortality, cardiovascular events, and kidney outcomes.

METHODS

This systematic review and meta‐analysis was conducted according to a prespecified protocol and in accordance with the Preferred Reporting Items for Systematic Reviews and Meta‐Analysis—2022 guidelines for conducting a meta‐analysis. ¹³ Data underlying this study are available in the article and provided in the results and tables. Any data will be freely shared from the corresponding author upon reasonable request.

Due to the study design, no institutional review board approval or patient informed consent were obtained.

Search Criteria, Data Sources, and Search Strategy

A literature review was performed to identify studies with a population with CKD who underwent a 24‐hour ABPM as follows: Population: Adult (≥18 years of age) patients with CKD. Intervention: 24‐hour ABPM to assess nocturnal BP dipping status. Comparators: Two comparisons were undertaken. The first was examining the risk of nondipping status in patients with CKD versus those without CKD. The second was comparing adverse outcomes in CKD patients by dipping status (dippers versus nondippers). Outcomes: The primary outcome was all‐cause mortality by dipping status in patients with CKD. Secondary outcomes included cardiovascular events and CKD progression. Cardiovascular events were defined as fatal or nonfatal hospitalization with myocardial infarction, congestive heart failure, stroke, revascularization procedure, peripheral vascular disease, and nontraumatic amputation. CKD progression was defined as ≥50% decline in estimated glomerular filtration rate, initiation of kidney replacement therapy, or reaching kidney failure. Methodology: Eligible studies included interventional trials (randomized, nonrandomized, or uncontrolled) and observational studies (prospective or retrospective cohorts, case–control, or cross‐sectional designs). Only full‐text, peer‐reviewed publications were included.

We searched MEDLINE, Embase, and Cochrane Central Registry of Controlled Trials from inception until December 1, 2023. Our search terms used a combination of keywords, Medical Subject Headings terms, and subject headings for CKD and ABPM. The search strategy is provided in Data S1. References were exported and managed through Zotero software.

Study Selection and Eligibility Criteria

Titles and abstracts in the initial search were independently screened by 2 authors (S.A., G.S.) based on the eligibility criteria. Eligible studies were included for independent full‐text review by the same authors. All discrepancies were resolved by consensus. We excluded case reports, case series, reviews, and studies addressing kidney transplant recipients, acute kidney injury, dialysis, and studies examining interventions (such as renal denervation) or drug effectiveness.

Data Extraction, Study Verification, and Quality Assessment

Data were extracted independently by 2 authors (S.A. and G.S.) using a standardized data extraction form. Discrepancies were resolved through consensus or with a third author (H.C.H.). The following data were extracted: study characteristics (journal, author, publication year, country, study design, duration of follow‐up), patient characteristics (age, sex, total number of patients), CKD definition, BP pattern (dipping versus nondipping), BP measuring device, and outcomes. Dippers were defined as a systolic BP (SBP) sleep‐awake ratio of 0.8 to 0.9 or a decline in nocturnal SBP ≥10%, otherwise, patients were defined as nondippers. Subgroups in nondippers included reverse dippers, defined as nighttime SBP reduction of 0%.

Risk of bias assessment was conducted by 2 reviewers (S.A. and G.S.) independently, with disagreements resolved through consensus. For the meta‐proportions analysis, we used the Joanna Briggs Institute Critical Appraisal Checklist for studies reporting prevalence data. ¹⁴ Studies were assigned an overall quality score ranging from 0 to 9, with scores of ≤5 categorized as low‐quality studies, 6 to 7 moderate quality, and 8 to 9 as high‐quality studies. For studies with a comparison group we used the Newcastle‐Ottawa Scale for observational studies. ¹⁵ Newcastle‐Ottawa Scale scores of ≥7 were considered high quality, 5 to 6 moderate quality, and <5 low quality.

Statistical Analysis

We conducted a meta‐proportional analysis and a meta‐analysis. The meta‐proportional analysis examined nondipping prevalence in the cohort with CKD. Tests of significance on the pooled proportions were calculated using the Exact fixed‐effects method, which provides standards and exact binomial CI without assuming homogeneity across study rates. ¹⁶ A sensitivity analysis was conducted, excluding studies with a high risk of bias such as: no clear CKD definition, those with a specific population with CKD (eg, IgA nephropathy or autosomal polycystic kidney disease only), studies without a clear definition of dipping status, studies that employed a dipping definition different from reduction in SBP >10% or daytime to nighttime SBP ratio 0.8 to 0.9 or studies with a Joanna Briggs Institute risk of bias score of ≤5. For the meta‐analysis, we used a dichotomous outcome with pooled risk ratios (RR) and a 95% CI. Random‐effects model was used for pooled estimates because of the heterogeneous nature of the included groups. Heterogeneity was examined using SD rates, inspecting the forest plots and I² statistics, with values of 25%, 50%, and 75% corresponding to low, moderate, and high levels of heterogeneity, respectively. Publication bias was assessed for analysis including more than eight studies visually by using funnel plots. For the meta‐analysis and generation of forest plots, we used Review Manager (Rev Man) 5.4 software (Cochrane Collaboration). All other statistical analysis were preformed using R statistical Software (version 4.5.2, Vienna, Austria). P<0.05 was deemed statistically significant.

RESULTS

Study Selection and Characteristics

After removal of duplicates, 2541 articles underwent initial screening, and 211 underwent full‐text screening. We identified 78 articles relevant to the research questions before excluding a further 15 articles that contained a patient population already included. ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²² , ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ Our final analysis included 63 studies published between 1996 and 2023 for the analysis of prevalence (Figure 1). Of note, 2 studies excluded from the prevalence analysis for containing similar patient cohorts were included in the outcome analysis since the initial included study did not include outcomes. ³² , ³³

Figure 1. Flow chart of study selection.

General characteristics of included studies are summarized in Table 1. A total of 54 515 patients were included in the analysis: 33825 with CKD and 20 690 with no CKD. Among the studies included 11 (18%) were retrospective, ³⁴ , ³⁵ , ³⁶ , ³⁷ , ³⁸ , ³⁹ , ⁴⁰ , ⁴¹ , ⁴² , ⁴³ , ⁴⁴ 24 (39%) prospective, ⁹ , ⁴⁵ , ⁴⁶ , ⁴⁷ , ⁴⁸ , ⁴⁹ , ⁵⁰ , ⁵¹ , ⁵² , ⁵³ , ⁵⁴ , ⁵⁵ , ⁵⁶ , ⁵⁷ , ⁵⁸ , ⁵⁹ , ⁶⁰ , ⁶¹ , ⁶² , ⁶³ , ⁶⁴ , ⁶⁵ , ⁶⁶ , ⁶⁷ 24 (39%) cross‐sectional, ⁶⁸ , ⁶⁹ , ⁷⁰ , ⁷¹ , ⁷² , ⁷³ , ⁷⁴ , ⁷⁵ , ⁷⁶ , ⁷⁷ , ⁷⁸ , ⁷⁹ , ⁸⁰ , ⁸¹ , ⁸² , ⁸³ , ⁸⁴ , ⁸⁵ , ⁸⁶ , ⁸⁷ , ⁸⁸ , ⁸⁹ , ⁹⁰ , ⁹¹ and 2 (3%) did not include a description. ⁹² , ⁹³ The smallest study included 21 patients with CKD, ⁹¹ and the largest included 6276 patients with CKD. ⁸⁴ The mean age of the participants ranged from 36 years ⁶⁵ to 74 years, ⁴³ with the proportion of men in the study ranging from 28% ⁷⁹ to 100%. ⁹²

Table 1.

General Characteristics of Included Studies

Author (y) [reference]	Country (number of centers)	Journal	Start to end date and follow‐up	Study design	Patient population/patient number	Men, n (%)	Age, y	CKD definition
Borrelli (2023) 9	Italy (3)	American Journal of Kidney Disease	2005–2015 7.8 y	Prospective	CKD/906	555 (61%)	ABPM goal Dipping: 58.7±15.0 Non‐dipping: 62.2±14.2 ABPM above goal Dipper: 65.5±13.1 Nondipper: 66.0±13.4	eGFR: <90
Agarwal* (2006) 32	United States (1 center)	Kidney International	June 2002–2004 2.5 y	Prospective	Diabetes, hypertension, glomerulonephritis, obstruction, and other/217	209 (96.3%)	67.4±10.9	NA
Wang* (2016) 33	China (1 center)	Scientific Reports	NA 2.9 y	Prospective	Chronic glomerulonephritis, diabetic nephropathy, hypertensive nephropathy, lupus nephritis, other/588	336 (57.1%)	42.76±16.71	CKD stages I–V according to Kidney Disease Improving Global Outcomes
Cha (2017) 34	Korea (4)	Journal of Hypertension	May 2013–October 2015 NA	Retrospective	Diabetic and nondiabetic nephropathy/351	222 (63%)	58 (mean)	CKD stages I–V
Ida (2019) 35	Japan (1)	Scientific Reports	October 2006–July 2016 4.7 y	Retrospective	Diabetic and nondiabetic nephropathy/1107	763 (69%)	70 [63–78]	eGFR: <60
Jacob (2004) 36	Germany (1)	Journal of Human Hypertension	January 1995–December 1996 3 y	Retrospective	Chronic GN, interstitial nephropathy, renal artery stenosis (>70%) with ischemic kidney disease, chronic pyelonephritis, obstructive hydronephrosis, and autosomal dominant polycystic kidney disease/95	61 (64%)	53.6±11.9	Confirmed with biopsy or radiographic imaging studies
Kado (2019) 37	Japan (1)	Japanese Society of Hypertension	October 2006–May 2014 NA	Retrospective	CKD/676	458 (68%)	68.4±11.7	eGFR: <60
Kansui (2014) 38	Japan (1)	Clinical and Experimental Hypertension	January 2006–March 2011 NA	Retrospective	CKD and dialysis patients/80 and 20	74 (74%)	61.1±1.5	Kidney damage† or eGFR <60 for ≥3 mo
Nakai (2016) 39	Japan (1)	Clinical and Experimental Hypertension	2009–2012 NA	Retrospective	CKD/124	71 (57%)	64±14	eGFR: <45
Sahutoglu (2018) 40	Turkey (1)	Clinical Nephrology	January 2012–December 2016 2.2 y	Retrospective	CKD/191	105 (55%)	59.7±12.4	eGFR: <90
Shafi (2012) 41	United States (1)	Journal of the American Society of Hypertension	NA NA	Retrospective	CKD/156	69 (44%)	Hypertension: 71.3±10.5 White coat hypertension: 70.5±12.8 Masked hypertension: 73.1±9.1 Normotension: 67.1±11.9	Kidney damage† or eGFR <60
Son (2020) 42	Korea (1)	Kidney Research and Clinical Practice	August 2014–January 2018 1 y	Retrospective	Diabetic nephropathy, glomerulonephritis, hypertensive nephropathy, cystic kidney disease, and tubulointerstitial disease/387	226 (58%)	61 [52, 69]	eGFR: <90
Tudor (2013) 43	Romania (1)	Romanian Journal of Diabetes Nutrition and Metabolic Diseases	January 2011–April 2013 NA	Retrospective	CKD/129	89 (69%)	74±15.56	NA
Li (2013) 44	China (1)	Clinical Nephrology	January 2007–December 2009 NA	Retrospective	Diabetic and nondiabetic nephropathy/214	108 (50%)	Diabetic: 62.9±14.5 Nondiabetic: 61.0±10.3	NA
Afsar (2012) 45	Turkey (1)	Journal of Diabetes and Its Complications	May 2007–May 2010 NA	Prospective	CKD and non‐CKD/216	173 (80%)	56.6±9.4	eGFR: <60 24‐h urine albumin ≥30 mg/d
Agarwal (2009) 46	United States (1)	American Journal of Nephrology	NA 8.7 y	Prospective	Diabetic nephropathy, hypertensive nephropathy, glomerulonephritis, Obstruction and other/322	309 (96%)	65±13	eGFR of <60 or urine PCR ≥0.22 g/g
Elung‐Jensen (2008) 47	Denmark NA	Nephrology Dialysis Transplantation	NA NA	Prospective	Glomerulonephritis, diabetic nephropathy, autosomal dominant polycystic kidney disease, and chronic nephropathy/34	21 (62%)	58 [23–79]	eGFR<60
Fedecostante (2014) 48	Italy (1)	Public Library of Science, One	NA NA	Prospective	CKD and non‐CKD/1805	1012 (56%)	CKD: 66.2±11.4 Non‐CKD: 53.6±13.2	eGFR <60 in at least 2 creatinine obtained 3–6 mo apart
Fukuda (2008) 49	Japan (1)	Hypertension	NA NA	Prospective	CKD/65	33 (51%)	47±18	Kidney damage† or GFR <60 for ≥3 mo
Gabbai (2012) 50	United States (11)	Clinical Journal of American Society of Nephrology	2002–2003 5 y	Prospective	CKD/617	382 (62%)	60.3±10.1	GFR: 20–65
Hornstrup (2018) 51	Denmark (1)	International Journal of Nephrology and Renovascular Disease	NA NA	Prospective	CKD and non‐CKD/126	105 (83%)	CKD: 62±3 Non‐CKD: 62±10	eGFR: 15–59
Iimuro (2015) 52	Japan (17)	Clinical and Experimental Nephrology	September 2007–December 2012 NA	Prospective	CKD/1075	678 (63%)	Women: 58.5±12.3 Men: 62.0±10.6	eGFR: 10–59
Ishikawa (2008) 54	Japan and United States (Jichi medical school registry)	Journal of Clinical Hypertension	January 1992–January 1998 3.4 y	Prospective	CKD and non‐CKD/811	311 (38%)	72.9±9.8	eGFR: <60
Katafuchi (2017) 55	Japan (1)	Hypertension and Circulatory Control	June 2009–May 2016 2.3 y	Prospective	Chronic glomerulonephritis, hypertensive nephropathy, diabetic nephropathy, and other/421	281 (67%)	69.4 [20–91]	CKD stages I–V
Kuczera (2017) 56	Poland (1)	Arterial Hypertension	NA NA	Prospective	Autosomal dominant polycystic kidney disease and IgA nephropathy/40	NA	49.0±14.7	eGFR: 15.0–59.9
Minutolo (2011) 57	Italy (4)	Archives of Internal Medicine	January 2003–December 2005 4.3 y	Prospective	Hypertensive nephropathy, diabetic nephropathy, glomerulonephritis, tubulointerstitial nephritis, autosomal dominant polycystic kidney disease, and other/436	2542 (58%)	65.1±13.6	GFR <60 or 60–90+urine protein >0.3 g/24 h, in 2 consecutive visits with an interval of 3 mo or more
Minutolo (2021) 58	Italy (3)	Nephrology Dialysis Transplantation	2005–2015 10.7 y	Prospective	Hypertensive nephropathy, diabetic nephropathy, glomerulonephritis, autosomal dominant polycystic kidney disease, and other/906	552 (61%)	Women: 63.3±15.0 Men: 64.1±13.4	CKD Stages 2–5 and not on dialysis/transplant
Minutolo (2015) 59	United States (NAV)	Hypertension	January 2006–December 2010 3.6 y	Prospective	Hypertensive nephropathy, diabetic nephropathy, glomerulonephritis, autosomal dominant polycystic kidney disease, and other/182	120 (66%)	65.6±13.4	eGFR: <90
Park (2024) 60	Korea (3)	The Japanese Society of Hypertension	2013–2018 4.8 y	Prospective	CKD/1061	566 (53%)	61 [52, 69]	Kidney damage† or eGFR<60
Rahman (2020) 61	United States (7)	Journal of American Society of Nephrology	2003–2012 6.72 y	Prospective	CKD/1502	841 (56%)	63.1±10.3	eGFR: 20–70
Redon (2010) 62	Spain (1)	Journal of Hypertension	June 1999–June 2005 3.7 y	Prospective	Nephroangiosclerosis, glomerulonephritis, autosomal dominant polycystic kidney disease, tubule‐interstitial nephritis, obstruction/79	47 (59%)	57±11	eGFR: <60
Tucker (1997) 63	United Kingdom (1)	Nephrology Dialysis Transplantation	NA NA	Prospective	Autosomal dominant polycystic kidney disease, Reflux nephropathy, biopsy‐proven glomerulonephritis, and other/85	37 (43%)	49±14	NA
Vagiona (2012) 64	Greece (NA)	Hippokratia	June 2006–June 2008 NA	Prospective	CKD/30	NA	NA	eGFR <15 patients scheduled to commence hemodialysis
Yu (2022) 65	China (1)	Journal of Clinical Medicine	NA NA	Prospective	Type 2 diabetic nephropathy or primary glomerular disease/1002	655 (65%)	Primary glomerular disease: 57.9±11.9 Diabetic: 57.3±10.5	Type 2 diabetes: Urine albumin >30 mg/gCr and eGFR <60 Primary glomerular disease: kidney damage or eGFR <60 for >3 mo
Zhou (2017) 66	China (1)	Clinical Nephrology	June 2013–December 2024 NA	Prospective	IgA nephropathy/375	195 (52%)	36.5±11.7	Confirmed IgA nephropathy on kidney biopsy
Ohashi (2012) 67	Japan (1)	Kidney Blood Pressure Research	NA NA	Prospective	CKD/40	33 (82%)	68.7±11.6	eGFR <60 for at least 3 mo; urinary PCR (≥300 mg/g) or structural kidney disease
Zhang (2019) 68	China (1)	Journal of Clinical Hypertension	July 2010–December 2017 NA	Prospective	Glomerulonephritis, diabetic and hypertensive nephropathy or other/1509	880 (58%)	44.6±16.2	Kidney damage† or eGFR <60 for ≥3 mo
Cerasola (1998) 69	Italy (NA)	American Journal of Nephrology	NA NA	Cross‐sectional	Autosomal dominant polycystic kidney disease/80	42 (53%)	Essential hypertension: 45.26±1.26 Essential hypertension nonrenal failure: 43.18±3.5 Renal failure 56.54±3.29	Autosomal dominant polycystic kidney disease, as per ultrasound diagnosis
Che (2017) 70	China (1)	Acta Cardiologica	September 2005–September 2007 NA	Cross‐Sectional	Secondary or primary glomerulonephritis or CKD of unknown cause/257	142 (55%)	51.3±14.8	Kidney damage† or eGFR <60 for ≥3 mo
Crespo (2013) 71	Spain (238)	Journal of Biological and Medical Rhythm Research	NA 5.4 y	Cross‐sectional	Hypertensive nephropathy/2659	1584 (59%)	64.9±13.2	eGFR: <60 Urine albumin ≥30 mg/24 h or urine ACR ≥30 mg/g on at least 2 occasions ≥3 mo apart
Di Daniele (2017) 72	Italy (1)	Atherosclerosis	NA NA	Cross‐sectional	CKD and non‐CKD/547	192 (35%)	CKD: 77.2±7.8 Non‐CKD: 64.3±16.1	eGFR: <60
Feldstein (2012) 73	Argentina (1)	Clinical and Experimental Hypertension	May 2009–June 2010 NA	Cross‐sectional	CKD/231	80 (35%)	Women: 68±10.5 Men: 64.5±14.7	Kidney damage † or eGFR <60 for ≥3 mo
Haruhara (2015) 74	Japan (1)	Hypertension Research	April 2010–March 2013 NA	Cross‐sectional	Biopsy‐proven glomerulonephritis or benign nephrosclerosis/138	92 (67%)	45±15	As confirmed on kidney biopsy
Jaques (2018) 75	Switzerland (1)	Blood Pressure Monitoring	NA NA	Cross‐sectional	CKD/69	45 (65%)	69 [62–75]	eGFR: 0–45
Li (2018) 76	China (1)	Continuing Nursing Education	NA NA	Cross‐sectional	IgA, hypertensive, diabetic and ischemic nephropathy. Chronic nephritis, renal arteriosclerosis, and other/400	161 (40%)	Control: 62.6±11.5 Normal casual BP: 61.9±14.0 High BP: 63.4±14.4	CKD stages 1–5
Manousopoulos (2021) 77	Greece (1)	Hypertension Research	May 2016–April 2019 NA	Cross‐sectional	CKD/137	102 (74%)	69.6±7.8	eGFR: 15–60 on 2 visits with an interval of 3 mo
Mizuno (2012) 78	Japan (1)	Blood Pressure Monitoring	NA NA	Cross‐sectional	CKD/104	49 (47%)	44±17	Kidney damage† or eGFR<60 for ≥3 mo
Mojón (2013) 79	Spain (38)	Journal of Biological and Medical Rhythm Research	2008–2018 6.3 y	Cross‐sectional	CKD and non‐CKD/10271	5505 (54%)	58.0±14.2	eGFR <60, urine albumin ≥30 mg/24 h Urine ACR ≥30 mg/g Cr on at least 2 occasions ≥3 mo apart
Mwasongwe (2020) 80	United States (44)	Clinical Journal of American Society of Nephrology	2000–2004 NA	Cross‐sectional	CKD and non‐CKD/561	157 (28%)	CKD: 64±10 Non‐CKD: 62±9	Urine ACR >30 mg/g or eGFR <60
Oh (2017) 81	Korea (2)	Journal of Korean Medical Sciences	August 2014–May 2015 NA	Cross‐sectional	Glomerulonephritis, diabetic and hypertensive nephropathy, or autosomal dominant polycystic kidney disease/387	226 (58%)	61 [20–75]	Kidney damage† or eGFR <60 for ≥3 mo
Cha (2014) 82	Korea (21)	Hypertension Research	October 2009–May 2011 NA	Cross‐sectional	Diabetic and hypertensive nephropathy, chronic glomerulonephritis, and autosomal dominant polycystic kidney disease/1317	828 (63%)	58 (median)	eGFR: 15–89
Paoletti (2005) 83	Italy (1)	American Journal of Kidney Disease	January 2000–2003 NA	Cross‐sectional	CKD and non‐CKD/296	198 (67%)	Controls: 54±10 CKD I–II: 57±14 CKD III–V: 58±11	Kidney damage† or eGFR <60 for ≥3 mo
Pluta (2015) 84	Poland (1)	Arterial Hypertension	September 2012–November 2014 NA	Cross‐sectional	Biopsy‐proven chronic glomerulonephritis, hypertensive nephropathy, autosomal dominant polycystic kidney disease, gouty nephropathy, nephrolithiasis unilateral nephrectomy due to pyonephrosis, trauma or tumor/120	60 (50%)	CKD: 56±11 Control: 54±11	CKD stage I–III
Sarafidis (2018) 85	Greece (Spanish Registry)	Journal of Hypertension	2009–May 2014 NA	Cross‐sectional	CKD and non‐CKD/16546	7777 (47%)	CKD: 64.8±12.3 Non‐CKD: 56.4±13.3	Kidney damage† or eGFR <60
Shi (2017) 86	China (1)	International Journal of Clinical and Experimental Medicine	August 2014–December 2015 NA	Cross‐sectional	CKD/476	237 (50%)	62.34±13.53	Kidney damage† or eGFR <60 for ≥3 mo
Song (2020) 87	China (1)	Journal of Clinical Hypertension	NA NA	Cross‐sectional	Glomerulonephritis, diabetic and hypertensive nephropathy, or other/823	493 (60%)	43.79 (mean)	Kidney damage† or eGFR <60 for ≥3 mo
Stróżecki (2021) 88	Poland (1)	Blood Pressure Monitoring	November 2015–April 2017 NA	Cross‐sectional	Biopsy proven glomerulonephritis, autosomal dominant polycystic kidney disease, interstitial nephritis, and hypertensive nephropathy/150	83 (55%)	CKD: 54.4±13.0 Non‐CKD: 55.4±9.8	eGFR: 30–90
Wen (2020) 89	China (1)	BioMed Central Nephrology	NA NA	Cross‐sectional	Primary glomerular disease/1178	633 (54%)	38.8±14.0	Kidney damage† or eGFR <60 for ≥3 mo
Woodell T. (2016) 90	USA (1)	Blood Pressure Monitoring	January 2012–June 2013 NA	Cross‐sectional	CKD and non‐CKD/334	110 (33%)	72±6	eGFR‐cystatin C or eGFR‐serum creatinine <60
Wu (2015) 91	China (1)	Kidney and Blood Pressure Research	November 2011–October 2014 NA	Cross‐sectional	Glomerulonephritis, diabetic and hypertensive nephropathy, and other/241	119 (49%)	CKD: 59.3±14.3 Non‐CKD: 57.3±14.7	Kidney damage† or eGFR <60
Andoh (2009) 92	Japan NAV	Hypertension Research	NA NA	Cross‐sectional	Glomerulonephritis and non‐CKD/45	19 (42%)	Nephrotic: 58.5±14.8. Nonnephrotic: 54.3±18	Nephrotic syndrome: Creatinine >1.5 mg/dL, urine protein >3.5 g/d, serum albumin <3.0 g/dL Hemoglobin A1C <6.0% and fasting serum glucose <110 mg/dL
Jeong (2020) 93	United States (1)	Physiological Reports	NA NA	NA	Hypertensive nephropathy, autosomal dominant polycystic kidney disease, glomerulonephritis, and other/32	32 (100%)	Dippers: 58.0±6.4 Nondippers: 55.0±7.6	eGFR of 60–89 with concomitant urine ACR of >30 mg/g or eGFR of 30–59
Schoina (2020) 94	Greece (1)	American Journal of Nephrology	NA NA	NA	Diabetic and nondiabetic CKD/96	58 (60%)	Diabetic: 69.44±9.73 Nondiabetic: 68.35±10.52	Chronic Kidney Disease Epidemiology Collaboration eGFR <90 and ≥15

Data expressed as mean±SD or median [interquartile range] unless specified otherwise. ABPM indicates ambulatory blood pressure monitoring; ACR, albumin to creatinine ratio; BP, blood pressure; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate, expressed in mL/min per 1.73 m²; MDRD, NA, not available; and PCR, protein to creatinine ratio.

^*Reference to study excluded from prevalence analysis (due to duplication of data set) but included in the outcome analysis.

^† Kidney damage for ≥3 mo, as confirmed by kidney biopsy or signs of kidney damage, such as proteinuria, microalbumin, abnormal urinary sediment, or abnormalities on imaging studies, with or without a decrease in eGFR as estimated by modification of diet in renal disease formula.

For the meta‐proportional analysis, 61 studies were included to determine the prevalence of nondippers in CKD, ⁹ , ³⁴ , ³⁵ , ³⁶ , ³⁷ , ³⁸ , ³⁹ , ⁴⁰ , ⁴¹ , ⁴² , ⁴³ , ⁴⁴ , ⁴⁵ , ⁴⁶ , ⁴⁷ , ⁴⁸ , ⁴⁹ , ⁵⁰ , ⁵¹ , ⁵² , ⁵³ , ⁵⁴ , ⁵⁵ , ⁵⁶ , ⁵⁷ , ⁵⁸ , ⁵⁹ , ⁶⁰ , ⁶¹ , ⁶² , ⁶³ , ⁶⁴ , ⁶⁵ , ⁶⁶ , ⁶⁷ , ⁶⁸ , ⁶⁹ , ⁷⁰ , ⁷¹ , ⁷² , ⁷³ , ⁷⁴ , ⁷⁵ , ⁷⁶ , ⁷⁷ , ⁷⁸ , ⁷⁹ , ⁸⁰ , ⁸¹ , ⁸² , ⁸³ , ⁸⁴ , ⁸⁵ , ⁸⁶ , ⁸⁷ , ⁸⁸ , ⁸⁹ , ⁹⁰ , ⁹¹ , ⁹² , ⁹³ and 17 studies examined the risk of nondipping status between patients with and without CKD.^* For the meta‐analysis, we included 9 studies, comparing the risk of adverse outcomes between dipping and nondipping status in patients with CKD.^† Risk of bias assessment for the prevalence study using the Joanna Briggs Institute assessment tool is summarized in Figure 2 and Table S1. We found 12 studies with a high risk of bias, 21 with a moderate risk of bias, and 28 with a low risk of bias. For studies with outcomes included in the meta‐analysis, risk of bias was assessed with the Newcastle‐Ottawa Scale (Table S2). Five studies had a score of >7, indicating good quality, and 4 studies a score of 5 to 6, indicating moderate quality, whereas no studies had a low‐quality score. Lower scores were due to poor representation of the exposed cohort and adequacy of the follow‐up of the cohorts.

Figure 2. Risk of bias assessment scores.

For each question, a yes response indicated a low risk of bias, a no response indicated a high risk of bias, and an uncertain response indicated some concerns.

Prevalence of Nondippers in Patients With CKD

The definition of dipping status was mostly consistent across studies, with a dipper defined as either a reduction in SBP of ≥10% at nighttime compared with daytime in 43 studies^‡ or nighttime to daytime SBP ratio of 0.8 to 0.9 in 15 studies.^§ Three studies ⁴⁰ , ⁵⁵ , ⁶³ did not have a definition for dippers. Furthermore, 44 (70%) of the studies included extreme dippers in the definition and 16 (25%) did not. Measurements of 24‐hour ABPM and definition of dipping per study are summarized in Table 2.

Table 2.

Description of ABPM Monitor Used, Dipping Definition and Breakdown of Dipping Status by CKD for Included Studies

Author (y) [reference]	CKD	No CKD	ABPM monitor	Dipping definition	Dippers	Dippers	Nondippers	Nondippers
			Time interval between measurements		CKD	No CKD	CKD	No CKD
Borrelli (2023) 9	906	NA	Spacelabs 90207 15 min 07:00–22:00, 30 min 22:00–7:00	Nighttime to daytime SBP ratio of <0.9†	270	NA	636	NA
Agarwal* (2006) 33	217	NA	Spacelabs 90207 monitor (Spacelabs Inc., Redmond, WA, USA) 20 min in the day (06:00–22:00), 30 min at night (22:00–06:00)	Night/day ratio of SBP <0.9†	43	NA	174	NA
Wang* (2016) 33	588	NA	TM‐2430 Monitor (A&D, Tokyo, Japan) 15 min in daytime, and 30 min at night‐time	Reduction in SBP of ≥10% at nighttime compared with daytime‡	226	NA	362	NA
Cha (2017) 34	351	NA	A&D TM‐2430 30 min 24 h	Mean SBP ratio night‐time to daytime ≤0.9‡	118	NA	233	NA
Ida (2019) 35	1107	NA	Fukuda Denshi FB‐270 30 min day, 60 min night	Average diurnal SBP 10% higher than average nocturnal SBP‡	154	NA	953	NA
Jacob (2004) 36	95	NA	Spacelabs 90239A 20 min day, 30 min night	Day to night MAP ratio ≤0.9‡	37	NA	58	NA
Kado (2019) 37	676	NA	Fukuda Denshi FB‐270 30 min 07:00–22:00, 60 min 22:00–07:00	Average diurnal SBP was not 10% higher than the average nocturnal SBP†	117	NA	559	NA
Kansui (2014) 38	63	17	TM‐2421 20 min 24 h	10%–20% decrease in SBP nighttime compared with daytime§	15	5	48	12
Nakai (2016) 39	124	NA	A&D TM‐2431 No interval data	10%–20% drop of waking SBP§	25	NA	99	NA
Sahutoglu (2018) 40	191	NA	Spacelab (unspecified) 15 min 07:00–22:00, 30 min 22:00–07:00	Not reported††	96	NA	95	NA
Shafi (2012) 41	156	NA	Welch Allyn 6100 30 min day, 60 min night	Night/day SBP ratio <0.9†	46	NA	110	NA
Son (2020) 42	387	NA	Suntech Oscar 2 & IEM Mobile‐O‐Graph 30 min 24 h	Nighttime to daytime SBP ratio of 0.8–0.9§	169	NA	218	NA
Tudor (2013) 43	129	NA	Spacelabs 90217 No interval data	Decline in the nocturnal blood pressure of >10%†	56	NA	73	NA
Li (2013) 44	214	NA	No device information 30 min 06:00–22:00, 60 min 22:00–06:00	SBP or DBP decrease by at least 10% during the night‡	34	NA	180	NA
Afsar (2012) 45	126	90	Spacelabs 90207 20 min 07:00–23:00, 30 min 23:00–07:00	Reduction of >10% (when compared with the daytime values) in the SBP or DBP levels at night†	69	63	57	27
Agarwal (2009) 46	179	143	Spacelabs 90207 20 min 06:00–22:00, 30 min 22:00–06:00	Sleep/wake systolic arterial pressure ratio of 0.9 or less‡	43	62	136	81
Elung‐Jensen (2008) 47	34	NA	A&D TM‐2421 15 min 07:00–23:00, 30 min 23:00–07:00	Nighttime reduction in both SBPand DBP >10%†	29	NA	5	NA
Fedecostante (2014) 48	270	1535	No device information No interval data	Change in mean SBPawake to sleep ≥10%‡	74	623	196	912
Fukuda (2008) 49	65	NA	Terumo ES‐H531 30 min 24 h	Nighttime MAP <90% of the daytime average continuously for ≥1 h†	41	NA	24	NA
Gabbai (2012) 50	617	NA	Space‐Labs 90207 30 min 24 h	Reduction in BP of >10% at nighttime compared with daytime†	124	NA	493	NA
Hornstrup (2018) 51	70	56	A&D (unspecified) 15 min 06:00–22:00, 30 min 22:00–06:00 or HealthSTATS BPRO 15 min 24 h	Nocturnal SBP decrease of >10%†	39	46	31	10
Iimuro (2015) 52	1075	NA	A&D TM‐2421 30 min 24 h	Patients with nocturnal BP change >10% and <20%||	500	NA	575	NA
Ishikawa (2008) 54	537	274	Nippon Colin ABPM‐630, A&D TM‐2421 or A&D TM‐2425 30 min 24 h	Sleep/awake SBP: 0.8–0.9§	324	184	213	90
Katafuchi (2017) 55	421	NA	Terumu ES‐H531: June 2009–November 2011 30 min day, 60 min night or Fukuda Denshi FM‐800: December 2011 onward 30 min day, 30 min night	Reduction at night in SBPbetween >10% and <20% compared with daytime||	85	NA	336	NA
Kuczera (2017) 56	40	NA	A&D TM‐2430 15 min 24 h	Not reported††	21	NA	19	NA
Minutolo (2011) 57	436	NA	Spacelabs 90207 15 min 07:00–22:00, 30 min 22:00–07:00	Night/day SBP 0.80 to less than 0.90¶	188	NA	248	NA
Minutolo (2021) 58	906	NA	Spacelabs 90207 15 min 07:00–23:00, 30 min 23:00–07:00	Day/night ratio of SBP <0.9†	270	NA	636	NA
Minutolo (2015) 59	136	NA	Spacelabs 90207 15 min 07:00–23:00, 30 min 23:00–07:00	Night/day ratio of SBP <0.90†	37	NA	99	NA
Park (2024) 60	1061	NA	A&D TM‐2430 30 min 24 h	Night/day SBP ratio of 0.8–0.9§	474	NA	587	NA
Rahman (2020) 61	1502	NA	Spacelabs 90207 or 90217 30 min 24 h	Night/day SBPratio <0.9†	559	NA	943	NA
Redon (2010) 62	79	NA	Spacelabs 90207 20 min 06:00–24:00, 30 min 24:00–06:00	>10% fall in the mean SBPor DBP recorded at night from the average daytime BP†	42	NA	37	NA
Tucker (1997) 63	85	NA	Spacelabs 90207 30 min 07:00–22:00, 60 min 22:00–07:00	Reduction in nighttime MAP >10% of daytime†	46	NA	39	NA
Vagiona (2012) 64	30	NA	No device information No interval data	Individuals who do not experience the normal decrease in BP during the night‡‡	9	NA	21	NA
Yu (2022) 65	1002	NA	No device information 15 min day, 30 min night	SBP reduction between 10% and 20%§	196	NA	806	NA
Zhou (2017) 66	375	NA	A&E (unspecified) 30 min 06:00–22:00, 60 min 22:00–06:00	Decline in nocturnal BP of ≥10% and <20%#	88	NA	287	NA
Ohashi (2012) 67	40	NA	No device information 20 min 06:00–22:00, 30 min 22:00–6:00	Decrease of >10% in systolic blood pressure during sleep compared with waking period†	14	NA	26	NA
Zhang (2019) 68	1509	NA	A&D TM‐2430 15 min day, 30 min night	Decline in nocturnal SBP >10%†	614	NA	895	NA
Cerasola (1998) 69	30	50	Spacelabs 90207 10 min 07:00–22:00, 15 min 22:00–07:00	Nocturnal reduction of mean BP of 10% or more of the average daytime BP‡	12	37	18	13
Che (2017) 70	257	NA	GE Tonoport V 30 min 06:00–22:00, 60 min 22:00–06:00	Night‐time blood pressure decreasing ≥10%‡	63	NA	194	NA
Crespo J. (2013) 71	2659	NA	Spacelabs 90207 20 min 07:00–23:00, 30 min 23:00–07:00 (48 h)	Decrease in mean BP during nighttime relative to during daytime of ≥10%‡	1014	NA	1645	NA
Di Daniele (2017) 72	224	323	Spacelabs 90207 or BP lab Petr Telegin Nizhny Novgorod 15 min 07:00–22:00, 30 min 22:00–07:00	Nighttime BP fall >10%†	39	181	185	142
Feldstein (2012) 73	231	NA	A&D TM‐2430 15 min 7:00–23:00, 30‐min 23:00–07:00	>10% fall in SBP during the nighttime period compared with daytime†	124	NA	107	NA
Haruhara (2015) 74	138	NA	A&D TM‐2431C 30 min day, 60 min night	Decrease in nocturnal BP of ≥10% and <20%#	58	NA	80	NA
Jaques (2018) 75	69	NA	Novacor SAS DIASYS INTEGRA II 15 min 07:00–22:00, 30 min 22:00–07:00	BPnighttime decline of at least 10% compared with daytime‡	29	NA	40	NA
Li (2018) 76	350	50	Spacelab 90207 No interval data	BPdrops >10% from the daytime†	151	45	199	5
Manousopoulos (2021) 77	137	NA	Spacelabs 90207 15 min 24 h	Nocturnal BP decrease >10%†	45	NA	92	NA
Mizuno (2012) 78	104	NA	Terumo ES‐H531 30 min 24 h	Nocturnal fall in MAP >10% from daytime to night‐time.†	32	NA	72	NA
Mojón (2013) 79	3227	7044	Spacelabs 90207 20 min 07:00–23:00, 30 min 23:00–07:00 (48 h)	Sleep‐time relative SBP decline >10%†	1271	4001	1956	3043
Mwasongwe (2020) 80	144	417	Spacelabs 90207 20 min 24 h	Night‐time to daytime decline in systolic blood pressure >10%†	34	140	110	277
Oh (2017) 81	387	NA	SunTech Medical Oscar 2 & IEM Mobile‐O‐Graph 30 min 24 h	Ratio nighttime to daytime SBP 0.8–0.9§	169	NA	218	NA
Cha (2014) 82	1317	NA	A&D TM‐2430 No interval data	Ratio of the mean SBP nighttime to daytime between 0.8 and 0.9§	635	NA	682	NA
Paoletti (2005) 83	244	52	Spacelabs 90207 30 min 08:00–20:00, 60 min 20:00–08:00	Nighttime averaged mean BP at least 10% lower than daytime averaged‡	52	36	192	16
Pluta (2015) 84	90	30	A&D TM‐2430 30 min 24 h	At least 10% drop in SBPand DBP‡	31	21	59	9
Sarafidis (2018) 85	6276	10270	Spacelabs 90207 20 min 24 h	>10% reduction in average nocturnal systolic blood pressure compared with day†	2485	5330	3791	4940
Shi (2017) 86	476	NA	Suntech Medical 30 min 06:00–22:00, 60 min 22:00–06:00	Nighttime BP reduction ≥10%‡	104	NA	372	NA
Song (2020) 87	823	NA	A&D TM‐2430 15 min day, 30 min night	Decline in the nocturnal SBP of ≥10%‡	255	NA	568	NA
Stróżecki (2021) 88	109	41	A&D TM‐2430 30 min 24 h	Night SBP fall >10% and ≤20% (SBP night/day ratio ≥0.8 and <0.9)**	42	28	67	13
Wen (2020) 89	1178	NA	No device data 15 min day, 30 min night	Reduction in SBP of ≥10% at nighttime compared with daytime‡	372	NA	806	NA
Woodell (2016) 90	60	274	Spacelabs 90217 20 min day, 60 min night	SBP diurnal variation of at least 10%‡	27	168	33	106
Wu (2015) 91	152	NA	Meditech ABPM‐04 20 min 07:00–22:00, 30 min 22:00–07:00	10%–20% reduction of SBP at night§	33	NA	119	NA
Andoh (2009) 92	21	24	A&D TM‐2425 15 min 07:00–21:00, 30 min 21:00–07:00	Sleeping/waking ratio of mean ambulatory BP ≤0.9‡	5	15	16	9
Jeong (2020) 93	32	NA	Spacelabs 90227 20 min day, 30 min night	Night time MAP average >10% lower than the daytime average†	8	NA	24	NA
Schoina (2020) 94	96	NA	IEM Mobil‐O‐Graph 20 min 07:00–23:00, 30 min 23:00–7:00	Fall of SBP at night >10% and ≤20%**	20	NA	76	NA

ABPM indicates ambulatory blood pressure monitoring; BP; blood pressure; CKD, chronic kidney disease; DBP, diastolic blood pressure; MAP, mean arterial pressure; NA, not available; and SBP, systolic blood pressure.

^*Reference to study excluded from prevalence analysis (due to duplication of data set) but included in the outcome analysis.

^† Dipping defined as >10% daytime systolic blood pressure (26 studies, 41% of included studies).

^‡ Dipping defined as ≥10% daytime systolic blood pressure (18 studies, 29% of included studies).

^§ Dipping defined as a decrease of 10% to 20% daytime systolic blood pressure (9 studies, 14% of included studies).

^|| Dipping defined as a decrease of >10% to <20% daytime systolic blood pressure (2 studies, 3% of included studies).

^¶ Dipping defined as a decrease of >10% to 20% daytime systolic blood pressure (1 study, 2% of included studies).

^# Dipping defined as a decrease of ≥10% to <20% daytime systolic blood pressure (2 studies, 3% of included studies).

^**Dipping defined as a decrease of >10% to ≤20% daytime systolic blood pressure (2 studies, 3% of included studies).

^†† Not reported (2 studies, 3% of included studies).

^‡‡ Other (1 study, 2% of included studies).

The pooled prevalence of nondippers in patients with CKD was 0.53 (95% CI, 0.52–0.53, SD=13.5%) (Figure 3). Across the studies examined, prevalence ranged from a low of 0.15 (95% CI, 0.06–0.30) ⁴⁷ to a high of 0.86 (95% CI, 0.84–0.88). ³⁵ Reverse dipping status in CKD was reported in 23 studies,^¶ with a pooled prevalence of 0.20 (95% CI, 0.19–0.20, SD=7.4%) ⁷³ and ranging from 0.12 (95% CI, 0.07–0.19) to 0.39 (95% CI, 0.35–0.43) ⁵⁰ (Figure 4). Funnel plots for nondipper and reverse dipper prevalence is presented in Figure S1.

Figure 3. Pooled prevalence of nondippers in patients with chronic kidney disease.

The squares represent the prevalence and the lines the 95% CIs. The diamond and the width represent the pooled prevalence and the 95% CIs.

Figure 4. Pooled prevalence of reverse dippers in patients with chronic kidney disease.

A sensitivity analysis was conducted, which excluded studies with no clear CKD definition, ⁴⁴ , ⁴⁵ , ⁶⁰ , ⁶² only a specific population with CKD, ⁶⁴ , ⁶⁵ , ⁶⁸ , ⁶⁹ , ⁸⁸ , ⁹¹ with an unclear or unconventional definition of dipping status, ⁴⁰ or studies with a high risk of bias.^** Thirty‐nine studies were included in the sensitivity analysis for the nondipper prevalence in CKD (prevalence, 0.53 [95% CI, 0.52–0.53], SD=14.2%; Figure S2), and 15 were included in the sensitivity analysis for reverse dipping (prevalence, 0.19 [95% CI, 0.18–0.20], SD=6.6%; Figure S3). Point estimates and heterogeneity tests were similar between the complete analysis and the sensitivity analysis.

Meta‐Analysis Comparing Nondipper Risk by CKD Status

Seventeen studies reported on nondipper status comparing patients with CKD to those with no CKD.^†† Patients with no CKD have a lower risk of nondipping status compared with patients with CKD (RR, 0.69 [95% CI, 0.63–0.75], SD=14.1%, P<0.001), with a high heterogeneity found across studies (I²=84%, P<0.001) (Figure 5). Funnel plot is present in Figure S4. In the sensitivity analysis, 14 studies were included showing a similar prevalence of 0.72 (95% CI, 0.66–0.78, SD=10%, P<0.001) (Figure S5) with a high heterogeneity (I²=83%, P<0.001).

Figure 5. Forest plot of the risk of nondipping status between patients with no CKD compared with patients with CKD.

The squares represent the risk ratio and the lines the 95% CIs for the individual studies. The area of each square is proportional to the study weight. The diamond and the width represent the pooled risk ratios and the 95% CIs. CKD indicates chronic kidney disease; and M‐H, Mantel–Haenszel.

Meta‐Analysis of Outcomes by Dipping Status in Patients With CKD

Nine studies examined outcomes in patients with CKD, comparing dippers to nondippers. For mortality, 4 studies (1016 dippers and 1727 nondippers) showed a significantly reduced risk of mortality in dippers compared with nondippers (RR, 0.60 [95% CI, 0.49–0.72], SD=3.2%, P<0.001) (Figure 6A), with a low heterogeneity among the studies (I²=0%, P=0.55). For cardiovascular events, 5 studies were included (1465 dippers and 2533 nondippers) with 1 study excluding patients with a history of cardiovascular events, ⁵³ 1 study excluding those with a cardiovascular event within 6 months of enrollment, ⁵⁰ and 3 studies including patients with a previous history of cardiovascular event. ⁹ , ⁵⁶ , ⁶⁰ A lower risk of cardiovascular events was observed in dippers compared with nondippers (RR, 0.63 [95% CI, 0.47–0.85], SD=28.3%, P<0.001) (Figure 6B) with moderate heterogeneity (I²=72%, P=0.006). Finally, 6 studies examined CKD progression (2089 dippers and 3425 nondippers), showing a significantly reduced risk of CKD progression in dippers compared with nondippers (RR, 0.64 [95% CI, 0.55–0.75], SD=14.1%, P<0.001) with low heterogeneity (I²=44%, P=0.09) (Figure 7). We examined subgroup differences between 3 kidney outcomes reported in the 6 studies: CKD progression only, kidney failure only, and both CKD progression and kidney failure. The risk of kidney outcome was similar across the subgroups with no significant subgroup differences (P=0.9) (Figure 7). No publication bias assessment was undertaken for any of the outcomes since the number of included studies for each outcome was <8.

Figure 6. Forest plot of the risk of (A) mortality and (B) cardiovascular events between patients with CKD with a dipping status and those with a nondipping status.

The squares represent the risk ratio and the lines the 95% confidence intervals for individual studies. The area of each square is proportional to the study weight. The diamond and the width represent the pooled risk ratios and the 95% CIs. CKD indicates chronic kidney disease.

Figure 7. Forest plot, including subgroup analysis, of the risk of CKD progression between patients with CKD with a dipping status and those with a nondipping status.

CKD indicates chronic kidney disease; KF, kidney failure; and M‐H, Mantel–Haenszel.

DISCUSSION

We conducted a systematic review and meta‐analysis examining the nondipping BP status in patients with CKD. First, we showed that a nondipping BP status is very common in the population with CKD, with a prevalence reaching 56%, whereas reverse dipping can affect almost a quarter of this cohort (23%). Second, we showed that CKD is a risk factor for nondipping BP status, compared with patients with no CKD, and can increase the risk by 30%. Finally, we examined adverse events associated with a nondipping status versus dipping status in patients with CKD and found a consistent and increased risk for mortality, cardiovascular events, and progression of CKD. To our knowledge, this is the first systematic review examining these associations.

Nondipping BP has been reported in 30% to 50% of the general population, with prevalence varying according to age, antihypertensive treatments, and comorbidities. ⁹⁴ Factors associated with a nondipping status include an older age, history of hypertension, and smoking. ⁹⁵ Chronic diseases are also a risk factors, with a high prevalence of nondipping status reported in patients with diabetes and cardiovascular disease. ⁹⁴ , ⁹⁶ , ⁹⁷ Although CKD has been shown to be a risk factor for nondipping BP status, our analysis adds to the literature by aggregating all available studies to present a unified point estimate, showing that the prevalence of nondippers in CKD can reach 60%. Our meta‐analysis comparison of cohorts with CKD to those without CKD also adds to the argument that CKD is a risk factor for nondipping status, with an almost 30% increased risk compared with patients without CKD.

The high prevalence of nondipping status in patients with CKD is not an unexpected result. In CKD, the loss of the normal circadian BP rhythm is likely driven by heightened sympathetic nervous system activity and impaired autonomic regulation, leading to elevated nocturnal BP. ⁹² Additionally, volume overload due to reduced sodium and water excretion in declining kidney function causes extracellular fluid expansion that blunts the normal nighttime BP decrease. ⁹⁸ These pathophysiological changes collectively contribute to the nondipping profile commonly observed in patients with CKD. However, it is important to clarify the extent of the prevalence, which is why a pooled analysis of the prevalence in the published literature is an important first step that needed to be undertaken.

We acknowledge the high heterogeneity in the prevalence analysis and the meta‐analysis of nondipping BP risk between patients with and without CKD. We attempted to examine the heterogeneity through a sensitivity analysis‐ excluding studies with, among other criteria, a high risk of bias, or studies with an incomplete or missing definition of CKD or method of nondipping measurement. Reassuringly the point estimate did not differ between the complete analysis and the sensitivity analysis, which strengthens the finding of high prevalence in this cohort of patient. However, the heterogeneity remained unchanged. There are other factors that can account for the high heterogeneity. CKD definition was not consistent across the studies or was very broad, with any patient with an estimated glomerular filtration rate<60 mL/min per 1.73 m² defined as having CKD in many studies. It is likely that some studies had a higher proportion of more advanced CKD than other studies, which may have accounted for the heterogeneity. Another factor to consider is the wide geographical distribution, with many countries and ethnicities accounted for in the pooled analysis. Finally, our included studies covered many eras from 1997 to 2023, and it is likely that increased recognition of hypertension management, treatment, and bedtime BP dosing may explain the difference across the years.

A high prevalence of nondippers in patients with CKD should be a point of concern. Studies in the people without CKD or the general population have documented that a nondipping status is associated with an increased risk of adverse events such as cardiovascular events, ⁹⁹ , ¹⁰⁰ incident CKD, ¹⁰¹ and mortality. ¹⁰² Our study adds to this knowledge by highlighting the high risk of mortality, cardiovascular events, and CKD progression in patients with CKD with a nondipping status, compared with patients with CKD with a normal dipping pattern. Heterogeneity was low or moderate in the adverse events outcome analysis, which contrasts with the prevalence meta‐proportional analysis. This supports the approach that regardless of the prevalence or the population studied, the presence of a nondipping BP status is consistently associated with a risk of adverse event.

Our analysis offers a significant contribution to understanding the burden and impact of nondipping BP in populations with CKD. We conducted a comprehensive search strategy to identify published articles, pooling a large number of participants (54 000) over many studies. We examined prevalence, comparison with patients without CKD, and adverse events, producing what should be a consistent and unified picture. We also included a wide selection of studies, which used a broad CKD definition and population group, which would increase the generalizability of our analysis.

We acknowledge the limitations of our analysis. As noted, we documented a high heterogeneity in the meta‐proportion analysis of prevalence and in the risk of nondipping status between CKD and non‐CKD. A subgroup analysis of nondipping prevalence by CKD stage may have explained some of the heterogeneity but was beyond the scope of this study. We also did not examine other forms of patients with kidney disorders such as those on kidney replacement therapy (hemodialysis, peritoneal dialysis, or kidney transplantation) or those with acute kidney injury. This will likely increase the heterogeneity of the presented results and would likely require a separate targeted analysis that was beyond the scope of this study. Finally, we would also have to caution against the conclusion that treating nondipping status in patients with CKD would result in a lower risk of adverse events, because there are no randomized controlled trials done to examine this intervention.

CONCLUSIONS

The study's large sample size, comparative design, and sensitivity analyses strengthen its conclusions; however, the high heterogeneity and variability in study definitions and quality limit the precision of pooled estimates. Despite these limitations, we believe that results of our analysis are clinically applicable and relevant. Nondipping status in CKD is highly prevalent and associated with adverse events. These conclusions suggest that an ABPM could be of clinical utility in patients with CKD. However, more important, it also highlights the need for prospective studies to determine if (1) medical intervention or therapy can improve the nondipping status in patients with CKD and (2) if controlling nondipping status in patients with CKD will result in a reduction in adverse events. We would strongly support considerations for randomized controlled trials given how common CKD is, how prevalent nondipping status is in patients with CKD, and the increased risk of adverse events seen in patients with CKD.

Sources of Funding

None.

Disclosures

None.

Supporting information

Data S1. Supplemental Methods

Tables S1 and S2

Figures S1–S5

PRISMA Checklist