Cath Lab Data Analysis

Author

Dr Samuel Blay Nguah

Published

April 19, 2024

Anlysis format

Demographic characteristics of participants
- Sex,
- Age,
- Marital status
- Nationality
- Ethnicity
- Race
- Religion
- Occupation
Type of procedures
- Which variable in the excel sheet determines this?
Prevalence of CVD risk factors
- Hypertension
- DM
- Dyslipidaemia
- Smoking
- Old age >/=60 yrs.
Let’s also look at proportion with multiple risk factors (1, 2, or 3 RF)
Types of coronary lesions (normal, single vessel, 2 vessel or 3 vessel disease)
- Which variables determines this in the excel sheet.
Coronary disease severity by risk level (no, one, two, three or more)
- Which variable in the excel sheet determines this?
Outcomes of PCI (emergency vs. elective)
- Which variable determines this
What is or are the outcome variables?
Mode of presentation(ACS vrs CCS)

Table 1: Demographics

Code

gtsummary::theme_gtsummary_compact()

Setting theme `Compact`

Code

table_1 <- 
    df_cathlab %>% 
    select(
        sex, age, marital_status, nationality,ethnicity, race, religion, 
        occupation
        ) %>% 
    gtsummary::tbl_summary(
        by = sex,
        digits = gtsummary::all_categorical()~ c(0, 1),
        statistic = gtsummary::all_categorical() ~ "{n}({p})",
        sort = list(everything() ~ "frequency"),
        missing = "no"
        ) %>% 
    gtsummary::bold_labels() %>% 
    gtsummary::modify_spanning_header(
        gtsummary::all_stat_cols() ~ "**Sex**"
        ) %>% 
    gtsummary::add_overall(last = T) %>% 
    gtsummary::modify_caption(
        caption = "**Table 1**: Demographic Charateristics of patients"
        ) 

table_1

**Table 1**: Demographic Charateristics of patients
Characteristic	Sex		Overall, N = 463¹
Characteristic	Female, N = 111¹	Male, N = 352¹	Overall, N = 463¹
Age (years)	64 (53, 71)	59 (52, 66)	60 (52, 67)
Marital Status
Married	63(66.3)	296(92.2)	359(86.3)
Single	16(16.8)	18(5.6)	34(8.2)
Widowed	12(12.6)	2(0.6)	14(3.4)
Divorced	4(4.2)	5(1.6)	9(2.2)
Nationaliity
Ghanaian	106(95.5)	307(87.2)	413(89.2)
Non-Ghanaian	5(4.5)	45(12.8)	50(10.8)
Ethnicity
Akan	47(43.9)	154(45.6)	201(45.2)
Others	18(16.8)	90(26.6)	108(24.3)
Ewe	23(21.5)	48(14.2)	71(16.0)
Ga-Adangbe	19(17.8)	46(13.6)	65(14.6)
Race
African	109(99.1)	331(95.7)	440(96.5)
Others	1(0.9)	15(4.3)	16(3.5)
Religion
Christian	103(95.4)	296(87.1)	399(89.1)
Islamic	5(4.6)	36(10.6)	41(9.2)
Others	0(0.0)	8(2.4)	8(1.8)
Occupation
Public Sector	26(24.1)	104(29.9)	130(28.5)
Retired	40(37.0)	77(22.1)	117(25.7)
Self-employed	24(22.2)	77(22.1)	101(22.1)
Private Sector	7(6.5)	89(25.6)	96(21.1)
Unemployed	11(10.2)	1(0.3)	12(2.6)
¹ Median (IQR); n(%)

Code

file.remove("table_1.docx")

[1] TRUE

Code

table_1 %>% 
    gtsummary::as_gt() %>% 
    gt::gtsave(filename = "table_1.docx")

Table 2: Clinicals

Code

table_2 <- 
    df_cathlab %>% 
    select(
        sex, presentation, urgency_of_procedure, type_of_intervention, 
        procedure, risk_factors_cat, type_of_lesion, urgency_of_procedure, 
        number_of_stents, in_hospital_outcome
        ) %>% 
    gtsummary::tbl_summary(
        by = sex,
        type = list(number_of_stents ~ "continuous"),
        digits = list(
            gtsummary::all_categorical()~ c(0, 1),
            number_of_stents ~ 0),
        statistic = gtsummary::all_categorical() ~ "{n}({p})",
        sort = list(c(
            in_hospital_outcome, urgency_of_procedure, 
            presentation) ~ "frequency"),
        missing = "no"
        ) %>% 
    gtsummary::bold_labels() %>% 
    gtsummary::modify_spanning_header(
        gtsummary::all_stat_cols() ~ "**Sex**"
        ) %>% 
    gtsummary::modify_caption(
        caption = "**Table 2**: Clinical Charateristics of patients"
        ) %>% 
    gtsummary::add_overall(last = T)

table_2

**Table 2**: Clinical Charateristics of patients
Characteristic	Sex		Overall, N = 463¹
Characteristic	Female, N = 111¹	Male, N = 352¹	Overall, N = 463¹
Presentation
CCS	85(76.6)	249(70.7)	334(72.1)
ACS	16(14.4)	67(19.0)	83(17.9)
Heart Failure	7(6.3)	33(9.4)	40(8.6)
Pre-op	3(2.7)	2(0.6)	5(1.1)
VHD	0(0.0)	1(0.3)	1(0.2)
Urgency of procedure
Elective	98(88.3)	289(82.1)	387(83.6)
Emergency	13(11.7)	63(17.9)	76(16.4)
Type of intervention
CABG	4(5.5)	22(8.6)	26(7.9)
OMT	31(42.5)	99(38.5)	130(39.4)
PCI	37(50.7)	134(52.1)	171(51.8)
POBA	1(1.4)	2(0.8)	3(0.9)
Procedure
CAG	74(66.7)	219(62.2)	293(63.3)
PCI	37(33.3)	133(37.8)	170(36.7)
No. of Risk factors
None	2(1.8)	15(4.3)	17(3.7)
1 to 3	74(66.7)	222(63.1)	296(63.9)
4 to 7	35(31.5)	115(32.7)	150(32.4)
Type of Coronary Lesion
Normal	36(32.4)	91(25.9)	127(27.5)
MNOD	22(19.8)	52(14.8)	74(16.0)
1VD	25(22.5)	65(18.5)	90(19.5)
2VD	15(13.5)	55(15.7)	70(15.2)
3VD	8(7.2)	62(17.7)	70(15.2)
LMD	5(4.5)	26(7.4)	31(6.7)
Number of Stents	1 (1, 2)	2 (1, 2)	2 (1, 2)
In-hospital Outcome
Discharged	110(99.1)	345(98.6)	455(98.7)
Died	1(0.9)	5(1.4)	6(1.3)
¹ n(%); Median (IQR)

Code

file.remove("table_2.docx")

[1] TRUE

Code

table_2 %>% 
    gtsummary::as_gt() %>% 
    gt::gtsave(filename = "table_2.docx")

Table 3: Clinicals

Code

table_3 <- 
    df_cathlab %>% 
    select(agecat, presentation,urgency_of_procedure, procedure, risk_factors_cat, 
           type_of_lesion, risk_factors_cat, in_hospital_outcome, 
           urgency_of_procedure, type_of_intervention, 
           number_of_stents) %>% 
    gtsummary::tbl_summary(
        by = agecat,
        type = list(number_of_stents ~ "continuous"),
        digits = list(
            gtsummary::all_categorical()~ c(0, 1),
            number_of_stents ~ 0),
        statistic = gtsummary::all_categorical() ~ "{n}({p})",
        sort = list(c(in_hospital_outcome, urgency_of_procedure, 
                      presentation) ~ "frequency"),
        missing = "no"
        ) %>% 
    gtsummary::bold_labels() %>% 
    gtsummary::modify_spanning_header(
        gtsummary::all_stat_cols() ~ "**Age in years**"
        ) %>% 
    gtsummary::modify_caption(
        caption = "**Table 3**: Clinical Charateristics of patients"
        ) %>% 
    gtsummary::add_overall(last = T)

17 observations missing `agecat` have been removed. To include these observations, use `forcats::fct_na_value_to_level()` on `agecat` column before passing to `tbl_summary()`.

Code

table_3

**Table 3**: Clinical Charateristics of patients
Characteristic	Age in years			Overall, N = 446¹
Characteristic	<45 years, N = 43¹	45 to 65 years, N = 273¹	>65 years, N = 130¹	Overall, N = 446¹
Presentation
CCS	30(69.8)	195(71.4)	96(73.8)	321(72.0)
ACS	8(18.6)	50(18.3)	22(16.9)	80(17.9)
Heart Failure	5(11.6)	26(9.5)	8(6.2)	39(8.7)
Pre-op	0(0.0)	2(0.7)	3(2.3)	5(1.1)
VHD	0(0.0)	0(0.0)	1(0.8)	1(0.2)
Urgency of procedure
Elective	37(86.0)	224(82.1)	112(86.2)	373(83.6)
Emergency	6(14.0)	49(17.9)	18(13.8)	73(16.4)
Procedure
CAG	30(69.8)	165(60.4)	89(68.5)	284(63.7)
PCI	13(30.2)	108(39.6)	41(31.5)	162(36.3)
No. of Risk factors
None	9(20.9)	8(2.9)	0(0.0)	17(3.8)
1 to 3	31(72.1)	188(68.9)	69(53.1)	288(64.6)
4 to 7	3(7.0)	77(28.2)	61(46.9)	141(31.6)
Type of Coronary Lesion
Normal	22(51.2)	72(26.5)	30(23.1)	124(27.9)
MNOD	6(14.0)	30(11.0)	33(25.4)	69(15.5)
1VD	9(20.9)	63(23.2)	16(12.3)	88(19.8)
2VD	2(4.7)	52(19.1)	14(10.8)	68(15.3)
3VD	3(7.0)	36(13.2)	27(20.8)	66(14.8)
LMD	1(2.3)	19(7.0)	10(7.7)	30(6.7)
In-hospital Outcome
Discharged	42(97.7)	267(98.2)	129(100.0)	438(98.6)
Died	1(2.3)	5(1.8)	0(0.0)	6(1.4)
Type of intervention
CABG	1(4.8)	15(7.6)	10(10.1)	26(8.2)
OMT	7(33.3)	71(36.0)	47(47.5)	125(39.4)
PCI	12(57.1)	109(55.3)	42(42.4)	163(51.4)
POBA	1(4.8)	2(1.0)	0(0.0)	3(0.9)
Number of Stents	2 (1, 2)	2 (1, 2)	2 (1, 3)	2 (1, 2)
¹ n(%); Median (IQR)

Code

file.remove("table_3.docx")

[1] TRUE

Code

table_3 %>% 
    gtsummary::as_gt() %>% 
    gt::gtsave(filename = "table_3.docx")

Table 4

Code

table_4 <- 
    df_cathlab %>% 
    select(type_of_lesion, risk_factors_cat, ethnicity) %>% 
    gtsummary::tbl_summary(
        by = type_of_lesion,
        digits = gtsummary::all_categorical()~ c(0, 1),
        statistic = gtsummary::all_categorical() ~ "{n}({p})",
        sort = list(gtsummary::everything() ~ "frequency"),
        missing = "no"
        ) %>% 
    gtsummary::bold_labels() %>% 
    gtsummary::modify_spanning_header(
        gtsummary::all_stat_cols() ~ "**Lesion Type**"
        ) %>% 
    gtsummary::modify_caption(
        caption = "**Table 3**"
        ) %>% 
    gtsummary::add_overall(last = T)

1 observations missing `type_of_lesion` have been removed. To include these observations, use `forcats::fct_na_value_to_level()` on `type_of_lesion` column before passing to `tbl_summary()`.

Code

table_4

**Table 3**
Characteristic	Lesion Type						Overall, N = 462¹
Characteristic	Normal, N = 127¹	MNOD, N = 74¹	1VD, N = 90¹	2VD, N = 70¹	3VD, N = 70¹	LMD, N = 31¹	Overall, N = 462¹
No. of Risk factors
1 to 3	76(59.8)	47(63.5)	69(76.7)	45(64.3)	39(55.7)	20(64.5)	296(64.1)
4 to 7	41(32.3)	26(35.1)	17(18.9)	24(34.3)	31(44.3)	10(32.3)	149(32.3)
None	10(7.9)	1(1.4)	4(4.4)	1(1.4)	0(0.0)	1(3.2)	17(3.7)
Ethnicity
Akan	58(46.0)	41(57.7)	40(47.1)	20(30.8)	29(43.3)	12(40.0)	200(45.0)
Others	25(19.8)	17(23.9)	24(28.2)	20(30.8)	13(19.4)	9(30.0)	108(24.3)
Ewe	26(20.6)	4(5.6)	14(16.5)	15(23.1)	11(16.4)	1(3.3)	71(16.0)
Ga-Adangbe	17(13.5)	9(12.7)	7(8.2)	10(15.4)	14(20.9)	8(26.7)	65(14.6)
¹ n(%)

Code

file.remove("table_4.docx")

[1] TRUE

Code

table_4 %>% 
    gtsummary::as_gt() %>% 
    gt::gtsave(filename = "table_4.docx")

Table 5

Code

table_5 <- 
    df_cathlab %>% 
    gtsummary::tbl_cross(
        row = urgency_of_procedure, 
        col = in_hospital_outcome,percent = "row",
        missing = "no") %>% 
    gtsummary::add_p() %>% 
    gtsummary::bold_labels()

FALSE observations with missing data have been removed.

Code

table_5

	In-hospital Outcome		Total	p-value¹
	Died	Discharged	Total	p-value¹
Urgency of procedure				<0.001
Elective	1 (0.3%)	384 (100%)	385 (100%)
Emergency	5 (6.6%)	71 (93%)	76 (100%)
Total	6 (1.3%)	455 (99%)	461 (100%)
¹ Fisher’s exact test

Code

file.remove("table_5.docx")

[1] TRUE

Code

table_5 %>% 
    gtsummary::as_gt() %>% 
    gt::gtsave(filename = "table_5.docx")

Graph of risk factors by sex

Code

df_temp <- 
    df_cathlab %>% 
    select(
        sex, hpn, t2dm, dyslip, prev_cad, smoking, obesity, alcohol, f_hx_cad, 
        phy_inactivity, old_age
        ) %>% 
    pivot_longer(cols = hpn:old_age) %>% 
    group_by(sex, name) %>% 
    count(value) %>% 
    pivot_wider(names_from = value, values_from = n) %>%  
    mutate(total = No + Yes)

df_temp2 <- 
    epiDisplay::ci.binomial(df_temp$Yes, df_temp$total) %>% 
    select(-total) %>% 
    cbind(df_temp) %>% 
    select(name, sex, events, total, probability, se, exact.lower95ci, exact.upper95ci) %>% 
    arrange(name, sex) %>% 
    mutate(
        name = case_when(
            name == "t2dm" ~ "Type II DM",
            name == "smoking" ~ "Smoking",
            name == "prev_cad" ~ "Previous CAD",
            name == "phy_inactivity" ~"Physical Inactivity",
            name == "old_age" ~ "Age >= 60 years",
            name == "dyslip" ~ "Dyslipidemia",
            name == "f_hx_cad" ~ "Family History of CAD",
            name == "alcohol" ~ "Alcohol use",
            name == "hpn" ~ "Hypertension",
            name == "obesity" ~ "Obesity",
            TRUE ~ name
            ) %>% factor(
                levels = c(
                    "Hypertension","Age >= 60 years", "Dyslipidemia", "Type II DM", 
                    "Obesity", "Previous CAD", "Physical Inactivity", "Alcohol use", 
                    "Family History of CAD", "Smoking")
                )
        )

df_temp2 %>% 
    mutate(across(probability:exact.upper95ci, ~round(.x*100, 1))) %>% 
    rename(
        "Prob" = "probability", 
        `Lower 95% CI` = exact.lower95ci,
        `Upper 95% CI` = exact.upper95ci) %>% 
    select(-c(events, total, se)) %>% 
    flextable::as_flextable() %>% 
    flextable::theme_vanilla() %>% 
    flextable::save_as_docx(path = "Table_for_Graph.docx")

df_temp2 %>% 
    ggplot(aes(y = probability, fill = sex, x = name)) +
    geom_bar(stat = "identity", position = position_dodge()) +
    geom_errorbar(
        aes(ymin = exact.lower95ci, ymax = exact.upper95ci),
        position = position_dodge(0.9),
        width = 0.25,
        linewidth = 0.2) +
    labs(fill = "Sex of Patient", y = "Proportion", x = NULL) +
    coord_flip() +
    theme_light() +
    theme(legend.position = "top")+
    scale_fill_brewer(palette = "Pastel1") +
    scale_y_continuous(limits = c(0, 1))

Code

 ggsave(filename = "Figure_1.jpg", width = 7, height = 5)

Table 6: Clinicals

Code

table_6 <- 
    df_cathlab %>% 
    select(risk_factors_cat, type_of_lesion, ethnicity) %>% 
    gtsummary::tbl_summary(
        by = ethnicity,
        digits = list(gtsummary::all_categorical()~ c(0, 1)),
        statistic = gtsummary::all_categorical() ~ "{n}({p})",
        missing = "no"
        ) %>% 
    gtsummary::bold_labels() %>% 
    gtsummary::modify_spanning_header(
        gtsummary::all_stat_cols() ~ "**Ethnicity**"
        ) %>% 
    gtsummary::modify_caption(
        caption = "**Table 6**: Clinical Charateristics of patients"
        ) %>% 
    gtsummary::add_overall(last = T)

18 observations missing `ethnicity` have been removed. To include these observations, use `forcats::fct_na_value_to_level()` on `ethnicity` column before passing to `tbl_summary()`.

Code

table_6

**Table 6**: Clinical Charateristics of patients
Characteristic	Ethnicity				Overall, N = 445¹
Characteristic	Akan, N = 201¹	Ewe, N = 71¹	Ga-Adangbe, N = 65¹	Others, N = 108¹	Overall, N = 445¹
No. of Risk factors
None	5(2.5)	2(2.8)	3(4.6)	6(5.6)	16(3.6)
1 to 3	124(61.7)	44(62.0)	41(63.1)	73(67.6)	282(63.4)
4 to 7	72(35.8)	25(35.2)	21(32.3)	29(26.9)	147(33.0)
Type of Coronary Lesion
Normal	58(29.0)	26(36.6)	17(26.2)	25(23.1)	126(28.4)
MNOD	41(20.5)	4(5.6)	9(13.8)	17(15.7)	71(16.0)
1VD	40(20.0)	14(19.7)	7(10.8)	24(22.2)	85(19.1)
2VD	20(10.0)	15(21.1)	10(15.4)	20(18.5)	65(14.6)
3VD	29(14.5)	11(15.5)	14(21.5)	13(12.0)	67(15.1)
LMD	12(6.0)	1(1.4)	8(12.3)	9(8.3)	30(6.8)
¹ n(%)

Code

file.remove("table_6.docx")

[1] TRUE

Code

table_6 %>% 
    gtsummary::as_gt() %>% 
    gt::gtsave(filename = "table_6.docx")

Saving data

Code

write_rds(x = df_cathlab, file = "cath_lab_data.rds")