Abstract
Background and Objectives:
Prospective data on presentation and outcomes of colorectal cancer (CRC) in Nigeria are limited; however, emergency presentation with advanced disease is thought common.
Methods:
Consecutive CRC patients presenting at six sites over 6 years were included. Risk factors for emergency presentation were evaluated using logistic regression methods. Overall survival (OS) was compared between emergent and elective patients using Kaplan-Meier methods and the log-rank test.
Results:
Of 535 patients, 30.7% presented emergently. Median age was 56 years, 55% were men, and 5.0% reported a cancer family history. Emergency patients had more proximal cancers (42.1% vs 24.0%), stage IV disease (61.6% vs 40.2%; p<0.001), lower household income (₦35,000/month vs ₦50,000/month), lower education levels (p=0.008) and accessed care with nonmotorized transport (50.6% vs 37.2%; p=0.005). Median OS was shorter in the emergency group (6.4 vs 17.4 months; p<0.001). Across clinical stages, emergency presentation was associated with worse OS (stage IV median OS 4.8 vs 9.4 months; p=0.002). Surgery improved survival in both groups, although emergency patients had higher 30-day postoperative mortality (23.2% vs 9.1%; p<0.001).
Conclusions:
Emergent Nigerian CRC patients have worse OS than elective patients. Cancer control efforts should focus on faster cancer detection, early presentation, diagnosis, and treatment.
Keywords: Colorectal cancer, Cancer survival, Emergency presentation, Nigeria, West Africa, Low- and middle-income countries (LMICs)
Summary for Table of Contents:
Colorectal cancer outcomes are poor in Nigeria compared to high-income countries. Emergency presentation in Nigerian colorectal cancer patients was associated with worse overall survival and higher 30-day postoperative mortality. Earlier detection of CRC is important to improve outcomes.
Introduction
Colorectal cancer (CRC) is the third most common cancer globally.1 CRC incidence and mortality are increasing in many low- and middle-income countries (LMICs) for which cancer surveillance data are available.2,3 In Nigeria and other sub-Saharan African countries, national population-based estimates of CRC incidence and mortality are lacking, although efforts are underway to improve cancer registration.4,5 Prospective facility data can also be a useful source of information for evaluating patients’ CRC presentation, treatment and survival outcomes.
In high-income countries (HICs), emergency presentation with - defined as a diagnosis of CRC that arose during an unplanned hospital admission and where CRC was the primary underlying reason for which admission was required -.has been associated with worse short-term6,7 and long-term8 outcomes, compared with elective presentation. In large administrative and registry database studies from North America,9 the UK,10 and Europe,11 18.5% to 23% of all patients with CRC presented to the hospital emergently. Single-center experiences in sub-Saharan African countries have generally reported much higher rates of emergency presentation; three small retrospective single-center Nigerian studies reported rates of 38%,12 45%,13 and 66%.14 The proportion of patients who present emergently with CRC is considered a performance indicator for health systems15,16 and may reflect delays in seeking care, reaching care, and referral; variation in the accessibility, quality, and timeliness of cancer care; and potential differences in the tumor biology. Emergency presentation with late-stage disease appears to be more common in countries that have low levels of population coverage of preventive, diagnostic, and treatment services for cancer.17 This pattern of presentation may contribute to the high mortality-to-incidence ratios observed for CRC in many LMICs.1,5 However, this effect has not been extensively characterized in Nigeria or other LMICs.
To better understand presentations and long-term outcomes of CRC in Nigeria, in 2013 the African Research Group for Oncology (ARGO) established a prospectively maintained database of consecutive patients with CRC who present to public hospitals in the Southwest region of Nigeria. Short- and long-term outcomes are determined through clinical follow-up and surveillance appointments and by dedicated telephone follow-up every 3 months with patients or, for those who have died, their nominated next of kin. All patients’ primary residences and hospital facilities are geolocated. Using this database, we aimed to identify demographic, socioeconomic, and clinical factors associated with emergency presentation with CRC, compared with elective presentation. We also sought to determine the effect of emergency presentation on short- and long-term outcomes, including overall survival (OS). The identification of factors that are associated with emergency presentation and the determination of their impact on survival after a diagnosis of CRC can help guide health providers and communities on how best to focus cancer control efforts and can inform improvements in cancer policy and health systems.
Methods
Study design, setting, and data collection
This study evaluated consecutive patients who presented emergently or electively with CRC in Nigeria. Data were obtained from the ARGO database, a prospectively maintained multisite database of six referral-level hospitals (Obafemi Awolowo University Teaching Hospital [OAUTH], Osun State; Federal Medical Centre, Owo, Ondo State; University of Ilorin Teaching Hospital, Ilorin, Kwara State; Ladoke Akintola University of Technology Teaching Hospital [since renamed University of Osun State Teaching Hospital], Osun State; Ondo State Trauma and Surgical Centre, Ondo State; and Lagos University Teaching Hospital, which provides cancer care across five states in South West and North Central Nigeria). The database was established through a partnership between Obafemi Awolowo University, Ile-Ife, Nigeria, and Memorial Sloan Kettering Cancer Center (MSK), New York, New York, USA, in 2013. Consecutive adult patients (aged >18 years) enrolled in the ARGO CRC database from April 2013 to May 2020 were included in this analysis. As only 1 patient was enrolled from the Lagos University Teaching Hospital site, this patient was included in the total counts, but demographic information was not detailed for the hospital site, to avoid potentially identifying the patient.
Data on patient, disease, and treatment variables were collected at the time of first presentation to a participating referral center. Factors that contributed to delays in diagnosis and information on both the optimal treatment recommended by the oncology team and the treatment the patient received were also collected.
Patient variables included age, sex, BMI, comorbidities, smoking status, highest education level, median household monthly income, and mode of transportation to the referral site (by foot, motorized, hybrid). Disease-related variables included mode of presentation (emergency vs elective), CRC site (cecum, ascending colon, transverse colon, descending colon, sigmoid, rectum), disease stage, intent of surgery (diagnostic, staging, curative, palliative), whether surgery was recommended and whether the patient underwent surgery, and the operation that was ultimately performed. The presence of preoperative perforation, bleeding, or obstruction was also captured. Disease stage was defined according to the eighth edition of the Union for International Cancer Control TNM Classification of Malignant Tumours. Disease stage is presented as clinical stage only, as access to biopsy and pathologic examination in Nigeria is variable and dependent on patient and hospital resources, especially for patients who do not undergo surgical resection, and therefore pathologic stage was often not available. Patient survival status was ascertained through clinic visits and follow-up phone calls every 3 months performed by a research assistant. In the case of patients who died during follow-up, next of kin were contacted to confirm the date of death and to ascertain, when available, the medically certified or presumed cause of death.
Data were entered into a secure online website using the REDCap system (Research Electronic Data Capture, Vanderbilt University, Nashville, TN), which is prospectively maintained and updated with each patient follow-up. The lead investigator and research coordinator at each hospital site verified the accuracy of all cases before submission.
Outcomes
Primary outcomes of interest were the mode of clinical presentation (emergency or elective) and OS. OS was calculated from the time of diagnosis to the time of death from any cause.
Statistical analysis
Demographic, clinical, treatment, and symptom characteristics were compared by presentation status using Fisher’s exact test for categorical variables and the Wilcoxon rank-sum test for continuous variables. Variables that were missing for >15% of patients were not formally compared. Patients who were alive at last follow-up were censored at the date of last follow-up. Kaplan-Meier methods were used to estimate OS from diagnosis. The log-rank test was used to compare OS by presentation and stage. Logistic regression was used to identify predictors of emergency presentation known at diagnosis.
Survival analyses were stratified by stage (stage I-II, stage III, and stage IV). Univariable Cox proportional hazards regression was used to assess the relationship between treatment and other covariates with OS. Time-dependent covariates were used for factors that occurred after the date of diagnosis, such as treatment. For formal analyses, surgery time was capped at 6 months; patients who had surgery after 6 months were not considered to have had timely and appropriate surgical management (note that the use of total neoadjuvant therapy in Nigeria, which may result in an appropriate delay of >6 months to the time of surgery, was not a treatment option during the study period). Landmarked Kaplan-Meier plots and estimates (at 3 months after diagnosis) were used to visualize the relationship between treatment and long-term OS and are presented in the Supplementary Material. Unless otherwise stated, unadjusted two-sided p<0.05 was considered statistically significant. All analyses were performed using SAS 9.4 (SAS Institute, Cary, NC).
Ethical approval
Ethical approval for enrollment with informed consent and maintenance of the ARGO prospective database was obtained from the Obafemi Awolowo University institutional review board and the MSK institutional review board (#18-114).
Results
A total of 535 consecutive patients with a diagnosis of CRC from six referral hospitals in Nigeria were registered in the ARGO database during the study period. Most patients (n=371 [69.3%]) presented electively to the hospital; the others presented emergently. The median age at presentation was 56 years (IQR, 43-65). More patients were men (n=294 [55%]), and most presented to OAUTH (n=314 [58.7%]). Emergency presentation was least common at OAUTH (n=70/314 [22.3%]) and most common at Federal Medical Centre, Owo (n=55/102 [53.9%]) (Table 1).
Table 1.
Demographic information for patients presenting with colorectal cancer in five Nigerian states, 2013-2020, by mode of presentation
| All Patients | Elective | Emergency | p- value |
||
|---|---|---|---|---|---|
| # Patients | 535 | 371 (69.3) | 164 (30.7) | ||
| Patient characteristics | |||||
| Age at Diagnosis, years | Median (IQR) | 56 (43-65) | 55 (44-65) | 56 (42-66) | 0.52 |
| N Missing | (0) | (0) | (0) | ||
| Age at Diagnosis (grouped) | 15-29 | 35 (6.5) | 22 (5.9) | 13 (7.9) | 0.35 |
| 30-44 | 114 (21.3) | 79 (21.3) | 35 (21.3) | ||
| 45-59 | 169 (31.6) | 125 (33.7) | 44 (26.8) | ||
| 60-74 | 172 (32.1) | 118 (31.8) | 54 (32.9) | ||
| 75+ | 45 (8.4) | 27 (7.3) | 18 (11) | ||
| Sex | Male | 294 (55) | 208 (56.1) | 86 (52.4) | 0.45 |
| Female | 241 (45) | 163 (43.9) | 78 (47.6) | ||
| Marital Status | Married | 414 (77.4) | 290 (78.2) | 124 (75.6) | 0.81 |
| Widowed | 63 (11.8) | 41 (11.1) | 22 (13.4) | ||
| Single | 47 (8.8) | 33 (8.9) | 14 (8.5) | ||
| Divorced | 11 (2.1) | 7 (1.9) | 4 (2.4) | ||
| Ethnicity | Yoruba | 487 (91) | 339 (91.4) | 148 (90.2) | 0.57 |
| Igbo | 23 (4.3) | 16 (4.3) | 7 (4.3) | ||
| Ibo | 4 (0.7) | 2 (0.5) | 2 (1.2) | ||
| Hausa | 3 (0.6) | 1 (0.3) | 2 (1.2) | ||
| Other | 18 (3.4) | 13 (3.5) | 5 (3) | ||
| Religion | Christian | 400 (74.8) | 274 (73.9) | 126 (76.8) | 0.45 |
| Muslim | 133 (24.9) | 96 (25.9) | 37 (22.6) | ||
| Unknown | 2 (0.4) | 1 (0.3) | 1 (0.6) | ||
| Hospital | OAUTHC | 314 (58.7) | 244 (65.8) | 70 (42.7) | <.001 |
| FMC Owo | 102 (19.1) | 47 (12.7) | 55 (33.5) | ||
| Ilorin | 91 (17) | 60 (16.2) | 31 (18.9) | ||
| Lautech | 19 (3.6) | 13 (3.5) | 6 (3.7) | ||
| OSTC Ondo | 8 (1.5) | 6 (1.6) | 2 (1.2) | ||
| LUTH | 1 (0.2) | 1 (0.3) | 0 (0) | ||
| Comorbidities | |||||
| Hypertension | Yes | 118 (22.1) | 90 (24.3) | 28 (17.1) | 0.07 |
| No | 415 (77.6) | 280 (75.5) | 135 (82.3) | ||
| Unknown | 2 (0.4) | 1 (0.3) | 1 (0.6) | ||
| Diabetes | Yes | 44 (8.2) | 32 (8.6) | 12 (7.3) | 0.73 |
| No | 489 (91.4) | 338 (91.1) | 151 (92.1) | ||
| Unknown | 2 (0.4) | 1 (0.3) | 1 (0.6) | ||
| HIV | Yes | 4 (0.7) | 3 (0.8) | 1 (0.6) | >0.95 |
| No | 529 (98.9) | 367 (98.9) | 162 (98.8) | ||
| Unknown | 2 (0.4) | 1 (0.3) | 1 (0.6) | ||
| Other | Yes | 18 (3.4) | 15 (4) | 3 (1.8) | 0.30 |
| No | 515 (96.3) | 355 (95.7) | 160 (97.6) | ||
| Unknown | 2 (0.4) | 1 (0.3) | 1 (0.6) | ||
| BMI (kg/m) | Median (IQR) | 22 (19-24) | 22 (19-24) | 21 (18-25) | |
| N Missing | (42) | (22) | (20) | ||
| Transportation (grouped) | Motorized Only | 235 (43.9) | 169 (45.6) | 66 (40.2) | 0.005 |
| Walk Only | 221 (41.3) | 138 (37.2) | 83 (50.6) | ||
| Combination | 79 (14.8) | 64 (17.3) | 15 (9.1) | ||
| Education Achieved | Secondary | 129 (24.1) | 84 (22.6) | 45 (27.4) | 0.008 |
| Graduate School | 109 (20.4) | 81 (21.8) | 28 (17.1) | ||
| Primary | 90 (16.8) | 55 (14.8) | 35 (21.3) | ||
| University | 160 (29.9) | 124 (33.4) | 36 (22) | ||
| None | 46 (8.6) | 26 (7) | 20 (12.2) | ||
| Unknown | 1 (0.2) | 1 (0.3) | 0 (0) | ||
| Household Salary | Median (IQR) | 45000 (19001-100000) | 50000 (19001-120000) | 35000 (15000-80000) | |
| N Missing | (321) | (232) | (89) | ||
| Family History of Cancer | Yes | 27 (5) | 16 (4.3) | 11 (6.7) | 0.28 |
| No | 508 (95) | 355 (95.7) | 153 (93.3) | ||
Advanced-stage disease at the time of presentation was common (Table 2). Only 4 patients overall presented with stage I disease. A higher proportion of patients who presented emergently had metastasis at the time of diagnosis (61.6% vs 40.2%; p<0.001), compared with patients who presented electively. Factors that contributed to patient-reported delays in presentation included reassurance from a health-care provider that symptoms were not concerning (26.8% emergency vs 19.7% elective), use of self-medication (7.3% emergency vs 11.9% elective), belief by the patient that their symptoms were not concerning (10.4% emergency vs 9.4% elective), and financial barriers (3.7% emergency vs 4.9% elective) (Table S1). Tumor location within the colon and rectum differed between the emergency and elective groups (Figure 1). Proximal tumors (proximal to the splenic flexure) were more common in the emergency group (42.1% emergency vs 24.0% elective), whereas rectal tumors were more common in the elective group (38.4% emergency vs 54.7% elective). Overall, 23.7% (n=63/266) of patients with rectal cancer and 38.7% (n=101/269) of patients with colon cancer presented emergently. Most patients were diagnosed primarily on the basis of clinical history, physical examination, imaging, and colonoscopy findings; however, in the emergency group, 28.7% of patients (n=47) were diagnosed on the basis of intraoperative findings during exploratory laparotomy (Table 2).
Table 2.
Clinical characteristics, work-up and treatment among patients presenting with colorectal cancer, by mode of presentation
| All Patients | Elective | Emergency | p- value |
||
|---|---|---|---|---|---|
| # Patients | 535 | 371 (69.3) | 164 (30.7) | ||
| Clinical T-Stage | T1 | 2 (0.4) | 2 (0.5) | 0 (0) | 0.005 |
| T2 | 10 (1.9) | 8 (2.2) | 2 (1.2) | ||
| T3 | 192 (35.9) | 148 (39.9) | 44 (26.8) | ||
| T4 | 321 (60) | 204 (55) | 117 (71.3) | ||
| Unknown | 10 (1.9) | 9 (2.4) | 1 (0.6) | ||
| Clinical N-Stage | N0 | 126 (23.6) | 96 (25.9) | 30 (18.3) | 0.029 |
| N1 | 230 (43) | 161 (43.4) | 69 (42.1) | ||
| N2 | 168 (31.4) | 104 (28) | 64 (39) | ||
| Unknown | 11 (2.1) | 10 (2.7) | 1 (0.6) | ||
| Clinical M-Stage | M0 | 257 (48) | 200 (53.9) | 57 (34.8) | <.001 |
| M1 | 250 (46.7) | 149 (40.2) | 101 (61.6) | ||
| Mx | 18 (3.4) | 13 (3.5) | 5 (3) | ||
| Unknown | 10 (1.9) | 9 (2.4) | 1 (0.6) | ||
| AJCC Stage | I | 4 (0.7) | 3 (0.8) | 1 (0.6) | <.001 |
| II | 98 (18.3) | 78 (21) | 20 (12.2) | ||
| III | 172 (32.1) | 131 (35.3) | 41 (25) | ||
| IV | 250 (46.7) | 149 (40.2) | 101 (61.6) | ||
| Unknown | 11 (2.1) | 10 (2.7) | 1 (0.6) | ||
| Abdominal Distention | Yes | 113 (21.1) | 44 (11.9) | 69 (42.1) | <.001 |
| No | 421 (78.7) | 326 (87.9) | 95 (57.9) | ||
| Unknown | 1 (0.2) | 1 (0.3) | 0 (0) | ||
| Abdominal Mass | Yes | 134 (25) | 93 (25.1) | 41 (25) | >0.95 |
| No | 400 (74.8) | 277 (74.7) | 123 (75) | ||
| Unknown | 1 (0.2) | 1 (0.3) | 0 (0) | ||
| Bleeding | Yes | 253 (47.3) | 193 (52) | 60 (36.6) | <.001 |
| No | 281 (52.5) | 177 (47.7) | 104 (63.4) | ||
| Unknown | 1 (0.2) | 1 (0.3) | 0 (0) | ||
| Obstruction | Yes | 148 (27.7) | 58 (15.6) | 90 (54.9) | <.001 |
| No | 387 (72.3) | 313 (84.4) | 74 (45.1) | ||
| Abdominal Pain | Yes | 275 (51.4) | 171 (46.1) | 104 (63.4) | <.001 |
| No | 259 (48.4) | 199 (53.6) | 60 (36.6) | ||
| Unknown | 1 (0.2) | 1 (0.3) | 0 (0) | ||
| Change in Bowel Habits | Yes | 325 (60.7) | 227 (61.2) | 98 (59.8) | 0.77 |
| No | 209 (39.1) | 143 (38.5) | 66 (40.2) | ||
| Unknown | 1 (0.2) | 1 (0.3) | 0 (0) | ||
| Weight Loss | Yes | 320 (59.8) | 225 (60.6) | 95 (57.9) | 0.57 |
| No | 214 (40) | 145 (39.1) | 69 (42.1) | ||
| Unknown | 1 (0.2) | 1 (0.3) | 0 (0) | ||
| Diagnosis Mode | Physical Exam | 184 (34.4) | 130 (35) | 54 (32.9) | <.001 |
| Colonoscopy | 118 (22.1) | 106 (28.6) | 12 (7.3) | ||
| Laparotomy | 68 (12.7) | 21 (5.7) | 47 (28.7) | ||
| CT Scan | 40 (7.5) | 29 (7.8) | 11 (6.7) | ||
| Clinical History | 7 (1.3) | 4 (1.1) | 3 (1.8) | ||
| Other/Multiple | 117 (21.9) | 81 (21.8) | 36 (22) | ||
| Unknown | 1 (0.2) | 0 (0) | 1 (0.6) | ||
| Abdominal CT | Yes | 346 (64.7) | 258 (69.5) | 88 (53.7) | <.001 |
| No | 137 (25.6) | 72 (19.4) | 65 (39.6) | ||
| Unknown | 52 (9.7) | 41 (11.1) | 11 (6.7) | ||
| Abdominal Chest | Yes | 145 (27.1) | 107 (28.8) | 38 (23.2) | 0.07 |
| No | 338 (63.2) | 221 (59.6) | 117 (71.3) | ||
| Unknown | 52 (9.7) | 43 (11.6) | 9 (5.5) | ||
| Chest X-Ray | Yes | 287 (53.6) | 201 (54.2) | 86 (52.4) | 0.32 |
| No | 191 (35.7) | 125 (33.7) | 66 (40.2) | ||
| Unknown | 57 (10.7) | 45 (12.1) | 12 (7.3) | ||
| Staging Laparoscopy | Yes | 5 (0.9) | 5 (1.3) | 0 (0) | 0.18 |
| No | 484 (90.5) | 324 (87.3) | 160 (97.6) | ||
| Unknown | 46 (8.6) | 42 (11.3) | 4 (2.4) | ||
| Staging Laparotomy | Yes | 288 (53.8) | 159 (42.9) | 129 (78.7) | |
| No | 179 (33.5) | 148 (39.9) | 31 (18.9) | ||
| Unknown | 68 (12.7) | 64 (17.3) | 4 (2.4) | ||
| Staging Physical Exam | Yes | 269 (50.3) | 172 (46.4) | 97 (59.1) | 0.79 |
| No | 15 (2.8) | 9 (2.4) | 6 (3.7) | ||
| Unknown | 251 (46.9) | 190 (51.2) | 61 (37.2) | ||
| Staging Ultrasound | Yes | 353 (66) | 246 (66.3) | 107 (65.2) | 0.33 |
| No | 136 (25.4) | 88 (23.7) | 48 (29.3) | ||
| Unknown | 46 (8.6) | 37 (10) | 9 (5.5) | ||
| Optimal Treatment | Surgery + Chemotherapy | 245 (45.8) | 155 (41.8) | 90 (54.9) | |
| Surgery + Chemotherapy + Radiotherapy | 162 (30.3) | 109 (29.4) | 53 (32.3) | ||
| Surgery | 37 (6.9) | 27 (7.3) | 10 (6.1) | ||
| Chemotherapy | 18 (3.4) | 15 (4) | 3 (1.8) | ||
| Chemotherapy + Radiotherapy | 5 (0.9) | 3 (0.8) | 2 (1.2) | ||
| Surgery + Chemotherapy + Observation | 4 (0.7) | 2 (0.5) | 2 (1.2) | ||
| Chemotherapy + Observation | 3 (0.6) | 3 (0.8) | 0 (0) | ||
| Surgery + Chemotherapy + Radiotherapy + Observation | 1 (0.2) | 1 (0.3) | 0 (0) | ||
| Surgery + Observation | 1 (0.2) | 0 (0) | 1 (0.6) | ||
| Unknown | 59 (11) | 56 (15.1) | 3 (1.8) | ||
| Received Chemotherapy | Yes | 229 (42.8) | 180 (48.5) | 49 (29.9) | |
| No | 151 (28.2) | 85 (22.9) | 66 (40.2) | ||
| Unknown | 155 (29) | 106 (28.6) | 49 (29.9) | ||
| Received Surgery | Yes | 346 (64.7) | 208 (56.1) | 138 (84.1) | <.001 |
| No | 122 (22.8) | 100 (27) | 22 (13.4) | ||
| Unknown | 67 (12.5) | 63 (17) | 4 (2.4) | ||
| RT Received | Yes | 6 (1.1) | 4 (1.1) | 2 (1.2) | |
| No | 286 (53.5) | 183 (49.3) | 103 (62.8) | ||
| Unknown | 243 (45.4) | 184 (49.6) | 59 (36) | ||
| Timing of Chemotherapy | Neoadjuvant | 128 (55.9) | 108 (60) | 20 (40.8) | 0.023 |
| Adjuvant | 99 (43.2) | 71 (39.4) | 28 (57.1) | ||
| Unknown | 2 (0.9) | 1 (0.6) | 1 (2) | ||
| Type of Chemotherapy | FOLFOX | 166 (75.5) | 135 (78) | 31 (66) | 0.11 |
| Xelox | 37 (16.8) | 28 (16.2) | 9 (19.1) | ||
| 5-FU+Leucovorin | 8 (3.6) | 4 (2.3) | 4 (8.5) | ||
| 5-FU | 3 (1.4) | 2 (1.2) | 1 (2.1) | ||
| Folfiri | 3 (1.4) | 1 (0.6) | 2 (4.3) | ||
| Imatinib | 2 (0.9) | 2 (1.2) | 0 (0) | ||
| Other/Multiple | 1 (0.5) | 1 (0.6) | 0 (0) | ||
Figure 1. Tumor location in patients who presented with colorectal cancer.
Distribution of tumor sites in the colon and rectum among patients who presented emergently or electively.
Socioeconomic characteristics of the patients are shown in Table 1. Patients who presented emergently had lower education levels (p=0.008), had a lower median household monthly income (₦35,000 [IQR, 15,000-80,000] vs ₦50,000 [IQR, 19,001-120,000]) and were more likely to have walked, rather than used motorized transport, to get to the hospital (50.6% vs 37.2%; p=0.005).
On univariable analysis, lower education level (p=0.009), hospital site (p<0.001), lower household income (p=0.047), and higher clinical stage (p<0.001) were associated with greater odds of emergency presentation; motorized transportation was associated with lower odds of emergency presentation (p=0.005). On multivariable analysis, factors that remained associated with higher odds of emergency presentation were stage IV disease (p=0.039), hospital site (p=0.002), and lower household income (p=0.032) (Tables S2 and S3).
The median follow-up for all surviving patients (n=202) was 16 months (IQR, 7-26); 15 months (IQR, 6-21) for the emergency group and 17 months (IQR, 8-36) for the elective group. OS was shorter in the emergency group (median OS, 6.4 months [95% CI, 4.8-9.5]) than in the elective group (median OS, 17.4 months [95% CI, 14.1-21.5]) (p<0.001, log-rank test; Table 3, Figure 2). When patients were stratified by clinical stage, OS was worse in patients who presented emergently than in patients who presented electively (stage I-II p=0.008, stage III p=0.003, stage IV p=0.002) (Table 3, Figure 3). Elective presentation and receipt of curative surgery were the only variables significantly associated with better survival across all clinical stages on univariable Cox regression analysis (Table S4). After adjusting for age at diagnosis, education achieved, sex, and transportation, emergency presentation remained associated with worse OS across all clinical stages while receipt of curative surgery remained associated with better OS for stage III and stage IV patients (Table 4). For patients with stage IV disease, education level and the mode of transportation to get to the hospital were also associated with OS, after adjustment for other covariates on multivariable analysis. Thirty-day postoperative mortality was higher in the emergency group (23.2%) than in the elective group (9.1%) (p<0.001, Table 3).
Table 3.
30 day perioperative mortality and overall survival by mode of presentation, clinical stage
| All Patients | Elective | Emergency | p-value | |
|---|---|---|---|---|
| # Patients | 535 | 371 (69.3) | 164 (30.7) | |
| Median OS, months (95% CI) | 13.3 (10.5-15.5) | 17.4 (14.1-21.5) | 6.4 (4.8-9.5) | <0.001 |
| 30 day perioperative mortality (%) | 51 (14.7) | 19 (9.1) | 32 (23.2) | <0.001 |
| Median OS by clinical stage, months (95% CI) | ||||
| Stage I/II | 34.3 (19.5-NR) | NR (24.0-NR) | 16.9 (5.4-48.8) | 0.008 |
| Stage III | 19.1 (13.9-28.2) | 26.8 (17.2-40.8) | 9.5 (3.7-16.4) | 0.003 |
| Stage IV | 7.0 (5.3-9.2) | 9.4 (6.9-11.5) | 4.8 (3.0-6.5) | 0.002 |
Notes: Log-rank p-values are comparing the distribution of OS for each group. OS = Overall survival; NR = Value not reached
Figure 2. Overall survival after diagnosis of colorectal cancer.
Kaplan-Meier survival curves show crude overall survival in 585 patients who presented with colorectal cancer to government hospitals in South South and South West Nigeria by mode of presentation (emergency vs elective) (A) and by stage at diagnosis (B). Median survival (months) and percentage of patients who died at 5 years, with 95% CIs, are presented as insets for each survival curve.
Figure 3. Overall survival after diagnosis of colorectal cancer among patients who presented emergently versus electively, by stage.
Kaplan-Meier survival curves show overall survival in patients who presented emergently or electively with stage I-II (A), stage III (B), or stage IV (C) colorectal cancer. Median survival (months) and percentage of patients who died at 5 years, with 95% CIs, are presented below each survival curve. OS, overall survival.
Table 4.
Multivariable Cox regression analysis assessing relationship between demographic, clinical, and treatment covariates with overall survival, by clinical stage
| I-II | III | IV | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| N(#E) | HR | [95% CI] | p-value | N(#E) | HR | [95% CI] | p-value | N(#E) | HR | [95% CI] | p-value | ||
| Age at Diagnosis, years | 1.02 | [0.99 - 1.05] | 0.13 | 0.99 | [0.98 - 1.01] | 0.45 | 1.00 | [0.99 - 1.01] | 0.75 | ||||
| Received Therapeutic Surgery | Yes | 35 (10) | 0.45 | [0.18 - 1.08] | 0.07 | 91 (37) | 0.41 | [0.26 - 0.65] | <.001 | 50 (35) | 0.44 | [0.29 - 0.66] | <.001 |
| No | 67 (29) | --- | 81 (55) | --- | 200 (166) | --- | |||||||
| Education Achieved | Graduate School | 24 (11) | 1.82 | [0.62 - 5.32] | 0.68 | 37 (20) | 0.57 | [0.30 - 1.08] | 48 (38) | 0.57 | [0.36 - 0.92] | ||
| Secondary | 18 (7) | 1.16 | [0.37 - 3.63] | 44 (25) | 0.52 | [0.28 - 0.96] | 63 (53) | 0.53 | [0.34 - 0.82] | ||||
| University | 33 (10) | 1.07 | [0.36 - 3.18] | 51 (20) | 0.47 | [0.25 - 0.88] | 70 (49) | 0.55 | [0.35 - 0.87] | ||||
| None | 9 (4) | 0.71 | [0.19 - 2.65] | 10 (5) | 0.80 | [0.30 - 2.15] | 0.15 | 27 (24) | 0.72 | [0.42 - 1.23] | 0.051 | ||
| Primary | 18 (7) | --- | 29 (21) | --- | 42 (37) | --- | |||||||
| Presentation | Emergency | 21 (13) | 3.21 | [1.52 - 6.79] | 0.002 | 41 (31) | 2.23 | [1.41 - 3.53] | <.001 | 101 (88) | 1.78 | [1.33 - 2.39] | <.001 |
| Elective | 81 (26) | --- | 131 (61) | --- | 149 (113) | --- | |||||||
| Sex | Male | 50 (17) | 1.18 | [0.58 - 2.40] | 0.65 | 97 (53) | 1.27 | [0.82 - 1.96] | 0.29 | 140 (111) | 1.03 | [0.76 - 1.38] | 0.86 |
| Female | 52 (22) | --- | 75 (39) | --- | 110 (90) | --- | |||||||
| Transportation | Combination | 16 (9) | 1.71 | [0.68 - 4.32] | 0.08 | 29 (12) | 1.24 | [0.61 - 2.54] | 0.84 | 30 (22) | 1.38 | [0.86 - 2.24] | 0.016 |
| Motorized Only | 45 (12) | 0.61 | [0.27 - 1.36] | 86 (50) | 1.07 | [0.67 - 1.71] | 99 (70) | 0.70 | [0.50 - 0.98] | ||||
| Walk Only | 41 (18) | --- | 57 (30) | --- | 121 (109) | --- | |||||||
Survival analysis with landmarking at 3 months was performed to elucidate the effects of treatment on survival among patients who did not die during the perioperative period and who received appropriate treatment (surgery or surgery plus chemotherapy) (Figures S1-S4). Receipt of curative-intent surgical resection +/− chemotherapy, where indicated, was associated with better survival, compared with lack of such treatment, for patients with stage I-III CRC and for both patients who presented emergently and patients who presented electively. However, survival for patients that received both surgery and chemotherapy remained markedly different between the emergency group (3-year OS, 37.3% [95% CI, 14.0%-61.1%], Figure S3) and the elective group (3-year OS, 70.3% [95% CI, 51.7%-82.8%], Figure S4). Among patients with stage IV disease, on landmarking at 3 months, receipt of surgical intervention (any surgical intervention) and chemotherapy was associated with an advantage in OS (median OS, 29.0 months [95% CI, 5.8 to not reached]), compared with receipt of surgery alone (median OS, 16.6 months [95% CI, 0.2-23.1]) or no surgery (median OS, 7.6 months [95% CI, 6.3-10.3]; p<0.001).
Discussion
Emergency presentation with advanced-stage disease is thought to be more common in countries without strong cancer control systems, contributing to worse short- and long-term outcomes. Nigeria lacks prospective studies with robust long-term follow-up data. In this study, emergency presentation with CRC strongly predicted worse short- and long-term OS, even after adjustment for demographic factors, tumor site and clinical stage. OS after a diagnosis of CRC in Nigeria was markedly lower than that in HICs, even in Nigerian patients who presented electively and had favourable disease characteristics. Low access to and uptake of recommended treatment - especially surgery and radiation - likely contributed to the observed poorer outcomes.
Some clinical and socioeconomic variables were associated with a higher risk of emergency presentation with CRC. Emergency presentation was associated with advanced-stage disease and occurred more often in patients with CRC proximal to the splenic flexure, similar to observations in other international experiences.18-20 For example, in a stage-adjusted analysis of >3000 patients in Ireland, metastatic disease and proximal tumor location were the two clinical factors associated with emergency presentation.18 Proximal tumors may have only vague symptoms or may remain asymptomatic until they cause intestinal obstruction or perforation, whereas even relatively early-stage rectal tumors can be symptomatic. Thus, presentation with advanced disease is more common for right-sided tumors. Right-sided tumors are less likely to be detected on fecal-based CRC screening and require full colonoscopy for screen-based detection. This is relevant for the design of CRC screening services. Tumors in the proximal colon exhibit different molecular characteristics and histologic profiles, with alternative pathways—namely, microsatellite instability and the CpG island methylation pathway—driving CRC tumorigenesis, presentation, and treatment responsiveness.21 There is evidence suggesting that these pathways may be more frequent in Nigeria and other West African countries.22 However, stage and anatomic site alone were not associated with the likelihood of emergency presentation in the present study or with OS in the emergency group.
Advanced-stage disease at presentation and emergency presentation have been associated with lower levels of access to care and with broader social and structural determinants of health.23-25 Low socioeconomic status and, to a lesser extent, low educational attainment have been consistently associated with emergency presentation of CRC in HICs, and this relationship persists after adjustment for age, cancer site, comorbidity, and availability of a regular source of primary care.10,23,25,26 In this study, emergency presentation was associated with lower median household income, lower education level, and reliance on nonmotorized transportation to access care. Lower socioeconomic status has previously been shown to be associated with a lower likelihood of cancer knowledge, use of screening,27 and adherence to cancer treatment28 in Nigeria, with the latter primarily a result of financial barriers to paying for care.28 In Nigeria, cancer costs are largely paid for out of pocket by the patient at the point of care,29 which may adversely affect stage at diagnosis, mode of presentation, and outcomes.30 Of note, in series from HICs, older adults and women are consistently more likely to present emergently.10,31 However, age and sex were not associated with mode of presentation or with survival in the present study.
The proportion of patients who present emergently with CRC can be considered a health system performance metric, reflecting cancer control efforts across prevention, screening, early detection, and treatment domains.16 The reported proportions of patients who present emergently in sub-Saharan Africa vary widely, with this proportion as high as two-thirds in some series.14,32 Previous small single-institution retrospective studies in Nigeria reported that 38%,12 45%,13 and 66%14 of patients with CRC presented emergently. Rates of emergency presentation in HICs have consistently decreased with the introduction of population-based CRC screening.33,34 This has occurred in concert with an increase in the incidence of early-stage disease at diagnosis.35 However, population-based CRC screening is not currently considered to be affordable, feasible, or cost-effective in most LMICs, including Nigeria.36,37 Strategies aimed at earlier detection and downstaging, including increasing the awareness of CRC symptoms and educating the public and health-care providers about CRC symptoms, could accelerate the time to diagnosis and treatment, without the resource intensity of population-based screening.37 Interestingly, the most common barrier to timely diagnosis reported in the present study, with equal frequency in the elective and emergency groups, was that a health provider had reassured the patient that the symptoms they reported (e.g., rectal bleeding) were not a cause for concern.
Emergency presentation with CRC has been consistently associated with worse short- and long-term survival in studies from HICs,11,12 even after adjustment for stage, tumor location, and other clinical and sociodemographic factors.18,28,38 In the present study, patients who presented emergently had worse 30-day postoperative mortality and OS. This was most pronounced in patients without metastases, even after stage of presentation and treatment received were accounted for and with landmarking the survival analysis at 3 months. Although the higher early (30-day) postoperative mortality can explain some of the worse outcomes among patients who presented emergently, these patients still had worse survival than patients who presented electively after treatment effects (surgery and chemotherapy) were accounted for and with landmarking survival at 3 months. Even among the most favorable prognostic group—that is, patients who presented electively with early-stage disease—the 3-year OS of 56% (95% CI, 41%-69%) observed in Figure 3A was worse than the ~90% OS seen in most HICs. Overall, patients who presented emergently were more likely to receive surgery (for diagnostic, therapeutic, or palliative intent) and less likely to receive chemotherapy, compared with patients who presented electively. Receipt of therapeutic surgery was associated with better OS in both groups. Despite presentation with curable disease, a significant proportion of patients did not receive surgery as recommended. This was more common among patients who presented electively—a group more likely to have outpatient management of their disease and a nonemergent indication for operative intervention. The reasons for this finding are not fully elucidated in the present study but may reflect the different acuity experienced by emergency patients, which often necessitates immediate surgical intervention at the time of presentation. Failure to initiate or complete treatment has been described—especially when health literacy is low and/or use of out-of-pocket payments is high—if patients leave the health facility after initial consultation, a situation more common when patients present electively.30 The most common reasons for not pursuing recommended treatment were patient death before receipt of treatment, cost of treatment, fear of colostomy, patient refusal, and treatment resources not being available (in particular, radiation therapy when recommended). The lack of receipt of recommended treatment is an area that requires further in-depth exploration, including with qualitative studies.
The differences in OS between patients who received treatment and those who did not, across stages of disease, suggest that substantial improvement in mortality could be made if all patients who present with curable disease, whether emergently or electively, receive recommended treatment. Patients with metastatic disease who underwent surgical intervention, including in the elective group, had better survival than those who did not undergo surgery. Selection bias may have contributed to this finding, with patients with better performance status more likely to be offered surgical intervention—a relationship that could not be accounted for in the present analysis. The finding of a treatment effect in these patients is at odds with observations in HICs, where removal of the primary tumor in the setting of incurable metastatic disease generally does not confer an OS advantage over treatment with systemic therapy.39 Among patients who presented emergently and received chemotherapy, chemotherapy was most frequently administered in the adjuvant setting. Because of the very low availability of radiation therapy, only ~1% of patients received radiation therapy (all such patients had rectal cancer, and there were no differences in receipt of radiation therapy between the emergency and elective groups). The lack of use of radiation therapy represents a very different treatment path than that taken for patients with rectal cancer in HICs, where neoadjuvant chemoradiation is the standard of care for locally advanced rectal cancer and where patients who present emergently are often diverted with surgery to facilitate chemoradiation, rather than undergoing an emergency resection.40
Short-term (<90 day) and postoperative (<30 day after surgery) survival are worse among patients who present emergently, compared with patients who present electively, in both HICs and LMICs, even when age, stage, and comorbidities are accounted for.41-43 In the present study, 30-day all-cause mortality was 23.2% in the emergency group versus 9.1% in the elective group. These rates are higher than those reported by the GlobalSurg3 study—a contemporary prospective cohort study of outcomes after CRC surgery in 84 countries that used a snapshot methodology (data collection across a 1- to 3-month period only). GlobalSurg3 reported a pooled 30-day postoperative mortality of 7.0% in LMICs, which included Nigeria.41 In the same study, postoperative mortality among elective patients (pooled across all country income groups) was 2.3%, compared with 10% among emergency patients. The higher postoperative mortality among patients who presented emergently can be explained, at least in part, by the higher likelihood that such patients were acutely physiologically unwell43—in particular, patients who presented with bowel obstruction or perforation, as performance status is often worse in patients with more-advanced-stage disease. We did not collect these data to adjust for clinical severity in our cohort; therefore, this explanation remains speculative. Future data collection in our prospective study will include measures of patient performance status and perioperative risk.
There are several limitations to this study. The six public tertiary hospitals included in the study were from an ethnically homogenous region of Nigeria (predominantly Yoruba) and may not be generalizable to other regions of the country. Also, cases that presented only to traditional and alternative health practitioners and smaller public and private hospitals were not captured. Thus, the database may have selected out a certain segment of the population that tended to use the public tertiary hospitals. We were also reliant on clinical stage, rather than pathologic stage, as the primary means for assigning stage at diagnosis, as only 64.7% of patients underwent surgery. In addition, access to imaging adjuncts like Chest CT and Chest Xray was variable. It is thus possible that distant and nodal disease were underreported, which may have contributed to the worse apparent survival at each reported stage regardless of presentation type, compared to high-income country results. We did not report CRC-specific survival, as data capture of this variable was challenging in a country where medically certified cause of death registration is not routine and where many patients did not have routine in-person or imaging-based surveillance to accurately identify and document recurrence. Finally, survival analysis assumes that censoring is uninformative, but losses to follow-up in this setting are often associated with other demographic, clinical, and socioeconomic variables that portend a worse outcome. With the acknowledgment of this limitation, our rates of long-term follow-up are some of the highest in any series from the region, and data quality and completeness have been a focus of the collaborative.
Conclusions
The incidence and outcomes of CRC are considered to be strong markers of the “cancer transition”—the process whereby countries undergoing rapid demographic and economic change transition from a cancer burden dominated by infection-related cancers to cancers associated with lifestyle risk factors more commonly in HICs.2,3 As countries transition, increasing incidence is typically accompanied by increasing mortality2 secondary to low levels of public and provider awareness and the reliance on health systems not optimized to diagnose and treat new patterns of disease.44 The sizeable proportion of patients with CRC in this study who presented emergently and with metastatic disease, as well as the poor 30-day and long-term survival outcomes in the setting of low access to and uptake of treatment, suggest that Nigeria is experiencing the first phase of this transition. In the absence of population-based cancer registries, prospectively collected hospital-based data) permit greater understanding of how patients with cancer present, what diagnostic workup and treatment they receive, and what clinical outcomes they experience. Perioperative optimization (even among patients who present emergently) and strengthening of postoperative care processes (including the capacity to rescue patients who experience complications) may help to address the high 30-day postoperative mortality reported in this study. Importantly, earlier detection, timely diagnosis, and improved access to and use of surgical, systemic, and radiation treatment would likely improve survival outcomes in the present study population.
Supplementary Material
Funding:
This work was supported, in part, by the National Institutes of Health (P30 CA008748) and the Thompson Family Foundation. The funders played no role in any aspect of this study.
Abbreviations:
- ARGO
African Research Group for Oncology
- CRC
colorectal cancer
- HICs
high-income countries
- LMICs
low- and middle-income countries
- OAUTH
Obafemi Awolowo University Teaching Hospital
- OS
overall survival.
Footnotes
Disclosures: None
References
- 1.Bray F, Ferlay J, Soerjomataram I, et al. : Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018;68(6):394–424. [DOI] [PubMed] [Google Scholar]
- 2.Arnold M, Sierra MS, Laversanne M, et al. : Global patterns and trends in colorectal cancer incidence and mortality. Gut 2017;66(4):683–691. [DOI] [PubMed] [Google Scholar]
- 3.Bray F, Soerjomataram I. The changing global burden of cancer: transitions in human development and implications for cancer prevention and control. In: Gelband H, Jha P, Sankaranarayanan R, Horton S, eds.: Cancer: Disease Control Priorities, Third Edition (Volume 3). Washington (DC): The International Bank for Reconstruction and Development/The World Bank; 2015. [PubMed] [Google Scholar]
- 4.Nigeria National Cancer Control Plan 2018-2022. Federal Ministry of Health, Government of Nigeria. https://www.iccp-portal.org/plans/nigeria-national-cancer-control-plan. Accessed June 19, 2024. [Google Scholar]
- 5.International Agency for Research on Cancer. Cancer in sub-Saharan Africa. In: Parkin DM, Ferlay J, Jemal A, et al. , eds.: IARC Scientific Publications No 167. Geneva, Switzerland: World Health Organization; 2018. [Google Scholar]
- 6.McPhail S, Elliss-Brookes L, Shelton J, et al. : Emergency presentation of cancer and short-term mortality. Br J Cancer 2013;109(8):2027–2034. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Brewster DH, Clark DI, Stockton DL, et al. : Characteristics of patients dying within 30 days of diagnosis of breast or colorectal cancer in Scotland, 2003-2007. Br J Cancer 2011;104(1):60–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Wanis KN, Ott M, Van Koughnett JAM, et al. : Long-term oncological outcomes following emergency resection of colon cancer. Int J Colorectal Dis 2018;33(11):1525–1532. [DOI] [PubMed] [Google Scholar]
- 9.Decker KM, Lambert P, Nugent Z, et al. : Time trends in the diagnosis of colorectal cancer with obstruction, perforation, and emergency admission after the introduction of population-based organized screening. JAMA Netw Open 2020;3(5):e205741. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Wallace D, Walker K, Kuryba A, et al. : Identifying patients at risk of emergency admission for colorectal cancer. Br J Cancer.2014;111(3):577–580. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gunnarsson H, Ekholm A, Olsson LI: Emergency presentation and socioeconomic status in colon cancer. Eur J Surg Oncol 2013;39(8):831–836. [DOI] [PubMed] [Google Scholar]
- 12.Ayandipo OO, Afuwape OO, Ojo AB, et al. : Perioperative morbidity and mortality after emergency and elective colon and proximal rectal surgery in ibadan. Ann Ib Postgrad Med 2020;18(1):24–30. [PMC free article] [PubMed] [Google Scholar]
- 13.Saluja S, Alatise OI, Adewale A, et al. : A comparison of colorectal cancer in Nigerian and North American patients: is the cancer biology different? Surgery. 2014;156(2):305–310. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Madubogwu C Colorectal cancer: late presentation and outcome of treatment. Afrimedic 2012;3(1):13–15. [Google Scholar]
- 15.Murchie P, Smith SM, Yule MS, et al. : Does emergency presentation of cancer represent poor performance in primary care? Insights from a novel analysis of linked primary and secondary care data. Br J Cancer 2017;116(9):1148–1158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.New Zealand Ministry of Health. Bowel Cancer Quality Performance Indicators: Descriptions. 2019. https://www.health.govt.nz/system/files/documents/publications/bowel-cancer-quality-performance-indicators-descriptions-mar19.pdf. Accessed July 22, 2024. [Google Scholar]
- 17.Martins T, Merriel S, Hamilton W. Routes to diagnosis of symptomatic cancer in sub-Saharan Africa: systematic review. BMJ Open 2020;10:e038605. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.McArdle CS, Hole DJ: Emergency presentation of colorectal cancer is associated with poor 5-year survival. Br J Surg 2004;91(5):605–609. [DOI] [PubMed] [Google Scholar]
- 19.McArdle CS, McMillan DC, Hole DJ: The impact of blood loss, obstruction and perforation on survival in patients undergoing curative resection for colon cancer. Br J Surg 2006;93(4):483–488. [DOI] [PubMed] [Google Scholar]
- 20.Oliphant R, Mansouri D, Nicholson GA, et al. : Emergency presentation of node-negative colorectal cancer treated with curative surgery is associated with poorer short and longer-term survival.: Int J Colorectal Dis 2014;29(5):591–598. [DOI] [PubMed] [Google Scholar]
- 21.Baran B, Mert Ozupek N, Yerli Tetik N, et al. : Difference between left-sided and right-sided colorectal cancer: a focused review of literature. Gastroenterology Res 2018;11(4):264–273. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Alatise O, Knapp GC, Sharma A, et al. : Novel biological insights into colorectal cancer from West Africa. Cancer Epidemiol Biomarkers Prev 2020;29(Suppl 12):IA35. [Google Scholar]
- 23.Pruitt SL, Davidson NO, Gupta S, et al. : Missed opportunities: racial and neighborhood socioeconomic disparities in emergency colorectal cancer diagnosis and surgery. BMC Cancer 2014;14:927. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Degett TH, Christensen J, Thomsen LA, et al. : Nationwide cohort study of the impact of education, income and social isolation on survival after acute colorectal cancer surgery. BJS Open 2020;4(1):133–144. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Askari A, Nachiappan S, Currie A, et al. : Who requires emergency surgery for colorectal cancer and can national screening programmes reduce this need? Int J Surg 2017;42:60–68. [DOI] [PubMed] [Google Scholar]
- 26.Douaiher J, Ravipati A, Grams B, et al. : Colorectal cancer-global burden, trends, and geographical variations. J Surg Oncol 2017;115(5):619–630. [DOI] [PubMed] [Google Scholar]
- 27.Sharma A, Alatise OI, O'Connell K, et al. : Healthcare utilisation, cancer screening and potential barriers to accessing cancer care in rural South West Nigeria: a cross-sectional study. BMJ Open 2021;11(7):e040352. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Sharma A, Alatise OI, Adisa AO, et al. : Treatment of colorectal cancer in sub-Saharan Africa: results from a prospective Nigerian hospital registry. J Surg Oncol 2020;121(2):342–349. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Twahir M, Oyesegun R, Yarney J, et al. : Access to care and financial burden for patients with breast cancer in Ghana, Kenya, and Nigeria. J Clin Oncol 2019;37:6562–6562. [Google Scholar]
- 30.Cazap E, Magrath I, Kingham TP, Elzawawy A: Structural barriers to diagnosis and treatment of cancer in low- and middle-income countries: the urgent need for scaling up. J Clin Oncol 2016;34(1):14–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Rabeneck L, Paszat LF, Li C: Risk factors for obstruction, perforation, or emergency admission at presentation in patients with colorectal cancer: a population-based study. Am J Gastroenterol 2006;101(5):1098–103. [DOI] [PubMed] [Google Scholar]
- 32.Herman AM, Hawkins AT, Misso K, et al. : Colorectal cancer in northern Tanzania: increasing trends and late presentation present major challenges. JCO Glob Oncol 2020;6:375–381. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Scholefield JH, Robinson MH, Mangham CM, Hardcastle JD: Screening for colorectal cancer reduces emergency admissions. Eur J Surg Oncol 1998;24(1):47–50. [DOI] [PubMed] [Google Scholar]
- 34.Libby G, Brewster DH, Steele RJ: Impact of faecal occult blood test screening on emergency admissions and short-term outcomes for colorectal cancer. Br J Surg 2014;101(12):1607–15. [DOI] [PubMed] [Google Scholar]
- 35.Fitzpatrick-Lewis D, Ali MU, Warren R, et al. : Screening for colorectal cancer: a systematic review and meta-analysis. Clin Colorectal Cancer 2016;15(4):298–313. [DOI] [PubMed] [Google Scholar]
- 36.Gelband H, Jha P, Sankaranarayanan R, Horton S, eds.: Cancer: Disease Control Priorities, Third Edition (Volume 3). Washington (DC): The International Bank for Reconstruction and Development/The World Bank; 2015. [PubMed] [Google Scholar]
- 37.Knapp GC, Alatise OI, Olasehinde OO, et al. : Is colorectal cancer screening appropriate in Nigeria? J Glob Oncol 2019;5:1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Weithorn D, Arientyl V, Solsky I, et al. : Diagnosis setting and colorectal cancer outcomes: the impact of cancer diagnosis in the emergency department. J Surg Res 2020;255:164–171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.van der Kruijssen DEW, Elias SG, Vink GR, et al. : Sixty-day mortality of patients with metastatic colorectal cancer randomized to systemic treatment vs primary tumor resection followed by systemic treatment: the CAIRO4 phase 3 randomized clinical trial. JAMA Surg 2021;156(12):1093–1101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Benson AB, Venook AP, Al-Hawary MM, et al. : NCCN Guidelines Insights: Rectal Cancer, Version 6.2020. J Natl Compr Canc Netw 2020;18(7):806–815. [DOI] [PubMed] [Google Scholar]
- 41.GlobalSurg Collaborative and National Institute for Health Research Global Health Research Unit on Global Surgery: Global variation in postoperative mortality and complications after cancer surgery: a multicentre, prospective cohort study in 82 countries. Lancet 2021;397(10272):387–397. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Biccard BM, Madiba TE, Kluyts HL, et al. : Perioperative patient outcomes in the African Surgical Outcomes Study: a 7-day prospective observational cohort study. Lancet 2018;391(10130):1589–1598. [DOI] [PubMed] [Google Scholar]
- 43.Degett TH, Dalton SO, Christensen J, et al. : Mortality after emergency treatment of colorectal cancer and associated risk factors-a nationwide cohort study. Int J Colorectal Dis 2019;34(1):85–95. [DOI] [PubMed] [Google Scholar]
- 44.Keum N, Giovannucci E: Global burden of colorectal cancer: emerging trends, risk factors and prevention strategies. Nat Rev Gastroenterol Hepatol 2019;16(12):713–732. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.