Age

Fig.1 Plot of Age distribution across Groups

The Dodged bar chart above represents individual counts representing frequency of age_grp categories 50-60,60-70,70-80 and 80-90 in categories BIPOLAR and THULEP belonging to group PROCEDURE_PERFORMED. Subgroup 60-70 has highest percentage 17/36 ( 47.22 % ) in group BIPOLAR . Subgroup 60-70 has highest percentage 21/36 ( 58.33 % ) in group THULEP . To formally check for association between groups we performed pearson chi-square test .

we foune found a Non-significant association between age_grp and PROCEDURE_PERFORMED. The chi-square statistic was 2.95 . The degree of freedom was 3 and P value was 0.4 .Contingency and Proportion table are shown below

Table 2

PROCEDURE_PERFORMED	age_grp	n	value	95 % Confidence Interval
BIPOLAR	50-60	1	1/27 ( 3.7 %)	0.4% - 16.04%
BIPOLAR	60-70	14	14/27 ( 51.85 %)	33.63% - 69.7%
BIPOLAR	70-80	12	12/27 ( 44.44 %)	27.06% - 62.95%
THULEP	50-60	3	3/25 ( 12 %)	3.5% - 28.67%
THULEP	60-70	15	15/25 ( 60 %)	40.58% - 77.25%
THULEP	70-80	6	6/25 ( 24 %)	10.69% - 42.94%
THULEP	80-90	1	1/25 ( 4 %)	0.44% - 17.21%

Table 3

	BIPOLAR	THULEP
50-60	1	3
60-70	14	15
70-80	12	6
80-90	0	1

Gender

Figure 2 CATHETERISATION Distribution in Our Population

The Dodged bar chart above represents individual counts representing frequency of CATHETERISED_OR_NOT categories 2 and 1 in categories BIPOLAR and THULEP belonging to group PROCEDURE_PERFORMED. Subgroup Yes has highest percentage 24/36 ( 66.67 % ) in group BIPOLAR . Subgroup Yes has highest percentage 22/36 ( 61.11 % ) in group THULEP . To formally check for association between groups we performed pearson chi-square test .

we we found a Non-significant association between CATHETERISATION and PROCEDURE_PERFORMED. The chi-square statistic was 0.06 . The degree of freedom was 1 and P value was 0.81 .Contingency and Proportion table are shown below

Table 4

PROCEDURE_PERFORMED	CATHETERISED_OR_NOT	n	value	95 % Confidence Interval
BIPOLAR	1	19	19/27 ( 70.37 %)	51.75% - 84.88%
BIPOLAR	2	8	8/27 ( 29.63 %)	15.12% - 48.25%
THULEP	1	13	13/25 ( 52 %)	33.1% - 70.46%
THULEP	2	12	12/25 ( 48 %)	29.54% - 66.9%

Table 5

	BIPOLAR	THULEP
1	19	13
2	8	12

RESECTION TIME

Figure 3 Boxplot Of Distribution Of RESECTION TIME in our Population

In this Figure we see Box plot of RESECTION_TIME_MIN in 2 sub-groups of PROCEDURE_PERFORMED : BIPOLAR and THULEP respectively .The individual jittered data points of RESECTION_TIME_MIN are overlaid over transparent Boxplot for better visualisation. We see distribution of data in individual sub-groups of PROCEDURE_PERFORMED based on these box-plots. The lower edge of box plot represents -first quartile (Q1), Horizontal bar represents the median, Upper edge represnts third quartile (Q3), Two black lines (whiskers) emanating from box-plots signify range of non-outlier data for the particular sub-group. Lower whisker represents minimum(Q1- 1.5 interquartile range) non-outlier limit of RESECTION_TIME_MIN and upper whisker represnts maximum(Q1+1.5interquartile range) of RESECTION_TIME_MIN .Any data beyond whiskers of box-plots represents outliers in the sub-groups The big brown point in the box-plots represents mean RESECTION_TIME_MIN of 2 groups and it has been annotated in the figure itself . Summary Statistics of the groups is presented in table below

Table 6 Summary Table Of RESECTION TIME within Groups

Group	n	Mean	SD	Median	Minimum	Maximum
BIPOLAR	27	89.593	11.949	86	70	116
THULEP	25	96.560	7.671	94	85	110

The mean in Group BIPOLAR [ 88.69 ± 11.83 ] was significantly lower than Group THULEP [ 96.11 ± 7.33 ] . The mean difference was -7.42 and 95 % confidence interval of the difference was ( -12.06 - -2.77 ) . The p value was <0.001 . The t statistic was -3.2 and degree of freedom of the Welch unpaired two-sample t test was 58.4 .In Formal statistical notation this result is expressed as : t(58.4) = -3.2, p= <0.001. The detailed statistical parameters of T test are given in table below.

TABLE 7

variable	group1	group2	statistic	df	p
RESECTION_TIME_MIN	BIPOLAR	THULEP	-2.52	44.7	0.02

BLOOD LOSS

Figure 3 Boxplot Of Distribution Of BLOOD LOSS in our Population

In this Figure we see Box plot of BLOOD_LOSS_DURING_PROCEDURE_LTR in 2 sub-groups of PROCEDURE_PERFORMED : BIPOLAR and THULEP respectively .The individual jittered data points of BLOOD_LOSS_DURING_PROCEDURE_LTR are overlaid over transparent Boxplot for better visualisation. We see distribution of data in individual sub-groups of PROCEDURE_PERFORMED based on these box-plots. The lower edge of box plot represents -first quartile (Q1), Horizontal bar represents the median, Upper edge represnts third quartile (Q3), Two black lines (whiskers) emanating from box-plots signify range of non-outlier data for the particular sub-group. Lower whisker represents minimum(Q1- 1.5 interquartile range) non-outlier limit of BLOOD_LOSS_DURING_PROCEDURE_LTR and upper whisker represnts maximum(Q1+1.5interquartile range) of BLOOD_LOSS_DURING_PROCEDURE_LTR .Any data beyond whiskers of box-plots represents outliers in the sub-groups The big brown point in the box-plots represents mean BLOOD_LOSS_DURING_PROCEDURE_LTR of 2 groups and it has been annotated in the figure itself . Summary Statistics of the groups is presented in table below

Table 8 Summary Table Of BLOOD LOSS within Groups

Group	n	Mean	SD	Median	Minimum	Maximum
BIPOLAR	27	0.505	0.161	0.452	0.32	0.878
THULEP	25	0.262	0.074	0.237	0.17	0.394

The mean Blood Loss in Group BIPOLAR [ 0.51 ± 0.17 ] was significantly higher than Group THULEP [ 0.28 ± 0.08 ] . The mean difference was 0.23 and 95 % confidence interval of the difference was ( 0.17 - 0.29 ) . The p value was <0.001 . The t statistic was 7.52 and degree of freedom of the Welch unpaired two-sample t test was 49.91 .In Formal statistical notation this result is expressed as : t(49.91) = 7.52, p= <0.001.

HOSPITAL STAY

Figure 5 Boxplot Of Distribution Of HOSPITAL STAY in our Population

In this Figure we see Box plot of HOSPITAL_STAY (in Hours) in 2 sub-groups of PROCEDURE_PERFORMED : BIPOLAR and THULEP respectively .The individual jittered data points of HOSPITAL_STAYHR are overlaid over transparent Boxplot for better visualisation. We see distribution of data in individual sub-groups of PROCEDURE_PERFORMED based on these box-plots. The lower edge of box plot represents -first quartile (Q1), Horizontal bar represents the median, Upper edge represnts third quartile (Q3), Two black lines (whiskers) emanating from box-plots signify range of non-outlier data for the particular sub-group. Lower whisker represents minimum(Q1- 1.5 interquartile range) non-outlier limit of HOSPITAL_STAYHR and upper whisker represnts maximum(Q1+1.5interquartile range) of HOSPITAL_STAYHR .Any data beyond whiskers of box-plots represents outliers in the sub-groups The big brown point in the box-plots represents mean HOSPITAL_STAY (Hour) of 2 groups and it has been annotated in the figure itself . Summary Statistics of the groups is presented in table below

Table 9 Summary Table Of Hospital stay within Groups

Group	n	Mean	SD	Median	Minimum	Maximum
BIPOLAR	27	38.815	6.816	40	26	48
THULEP	24	28.583	7.649	26	18	48

The mean Hospital Stay in Group BIPOLAR [ 38.42 ± 6.78 ] was significantly higher than Group THULEP [ 29.06 ± NA ] . The mean difference was 9.36 and 95 % confidence interval of the difference was ( 6.21 - 12.51 ) . The p value was <0.001 . The t statistic was 5.92 and degree of freedom of the Welch unpaired two-sample t test was 69 .In Formal statistical notation this result is expressed as : t(69) = 5.92, p= <0.001. The detailed statistical parameters of T test are given in table belo

CATHETRISATION DURATION (IN HOURS)

Figure 5 Boxplot Of Distribution Of CATHETARISATION DURATION in our Population

In this Figure we see Box plot of CATHETERISATION_DURATION post operatively in Hours in 2 sub-groups of TURP : BIPOLAR and THULEP respectively .The individual jittered data points of CATHETERISATION_DURATION are overlaid over transparent Boxplot for better visualisation. We see distribution of data in individual sub-groups of PROCEDURE_PERFORMED based on these box-plots. The lower edge of box plot represents -first quartile (Q1), Horizontal bar represents the median, Upper edge represnts third quartile (Q3), Two black lines (whiskers) emanating from box-plots signify range of non-outlier data for the particular sub-group. Lower whisker represents minimum(Q1- 1.5 interquartile range) non-outlier limit of CATHETERISATION_DURATION_POST_OP_HR and upper whisker represnts maximum(Q1+1.5interquartile range) of CATHETERISATION .Any data beyond whiskers of box-plots represents outliers in the sub-groups The big brown point in the box-plots represents mean CATHETERISATION_DURATION(in Hours) of 2 groups and it has been annotated in the figure itself Summary Statistics of the groups is presented in table below

Table 10 Summary Table Of Catheterisation Duration within Groups

Group	n	Mean	SD	Median	Minimum	Maximum
BIPOLAR	27	31.111	6.612	34	18	40
THULEP	23	20.522	6.266	19	12	36

The mean Catheterisation Duration in Group BIPOLAR [ 30.5 ± 6.4 ] was significantly higher than Group THULEP [ 21.26 ± NA ] . The mean difference was 9.24 and 95 % confidence interval of the difference was ( 6.39 - 12.08 ) . The p value was <0.001 . The t statistic was 6.49 and degree of freedom of the Welch unpaired two-sample t test was 67.41 .In Formal statistical notation this result is expressed as : t(67.41) = 6.49, p= <0.001.

Figure showing Overall Correlation between PROSTATE SIZE AND BLOOD LOSS

The scatter plots above show relationship between PROSTATE_SIZE on X axis and BLOOD_LOSS_DURING_PROCEDURE_LTR on Y axis in overall population. Graphically, we see that as PROSTATE_SIZE increases, BLOOD_LOSS_DURING_PROCEDURE mostly remains constant with a flat to negative slope. On a formal statistical linear regression analysis, we that line of best fit (blue line signifying line with least square difference) also has a negative slope implying a negative correlation. The gray shaded error around blue line signifies 95% confidence interval of linear regression line of best fit. The correlation between two variables is Non-Significant . The Pearson’s correlation between PROSTATE_SIZE and BLOOD_LOSS_DURING_PROCEDURE_LTR is -0.13 with 95% Confidence Interval of -0.35 to 0.11. the t statistic is -1.06 The p value is 0.29 .The degree of freedom is 70. In formal statistical notation this expressed as t(70)= -1.06, P= 0.29. r(Pearson) = -0.13 95% C.I. [-0.35-0.11]. n= 72. The correlation is summmarised in table below

We see that there is large variation between data points particualarlyy in prostate size range of 60-80 grams

However we see distinct variation between relationship between Blood loss and Prostate size in Groups if we see Group Wise. In Thulep Group there is significant negative correlation while curve stays flat in BIPOLAR Group.

Table 11. Table Summarizing Overall correlation between PROSTATE SIZE AND BLOOD LOSS

Group 1	Group 2	Degree of Freedom	T statistic	Correlation	95 % C.I.	P value
PROSTATE_SIZE	BLOOD_LOSS_DURING_PROCEDURE_LTR	50	-0.89	-0.13	-0.39-0.15	0.38

Table 12. Table Summarizing Overall correlation between PROSTATE SIZE AND BLOOD LOSS In THULEP SUBGROUP

Group 1	Group 2	Degree of Freedom	T statistic	Correlation	95 % C.I.	P value
PROSTATE_SIZE	BLOOD_LOSS_DURING_PROCEDURE_LTR	23	-2.94	-0.52	-0.76–0.16	0.01

Table 13. Table Summarizing Overall correlation between PROSTATE SIZE AND BLOOD LOSS In BIPOLAR SUBGROUP

Group 1	Group 2	Degree of Freedom	T statistic	Correlation	95 % C.I.	P value
PROSTATE_SIZE	BLOOD_LOSS_DURING_PROCEDURE_LTR	25	-0.43	-0.08	-0.45-0.3	0.67

Table 14 Table with summary statistics of PROSTATE SIZE AND BLOOD LOSS

variable	n	Mean	SD	Median	Minimum	Maximum
BLOOD_LOSS_DURING_PROCEDURE_LTR	52	0.388	0.175	0.364	0.17	0.878
PROSTATE_SIZE	52	80.442	15.897	78.000	60.00	130.000

Figure showing Overall Correlation between RESECTION TIME AND BLOOD LOSS

The scatter plots above show relationship between RESECTION_TIME in minutes on X axis and BLOOD_LOSS_DURING_PROCEDURE_LTR on Y axis. Graphically, we see that as RESECTION_TIME_MIN increases, BLOOD_LOSS_DURING_PROCEDURE_LTR decreases . On a formal statistical linear regression analysis, we that line of best fit (blue line signifying line with least square difference) also has a negative slope implying a negative correlation. The gray shaded error around blue line signifies 95% confidence interval of linear regression line of best fit. The correlation between two variables is Significant . The Pearson’s correlation between RESECTION_TIME_MIN and BLOOD_LOSS_DURING_PROCEDURE_LTR is -0.33 with 95% Confidence Interval of -0.53 to -0.11. the t statistic is -2.97 The p value is <0.001 .The degree of freedom is 70. In formal statistical notation this expressed as t(70)= -2.97, P= <0.001. r(Pearson) = -0.33 95% C.I. [-0.53–0.11]. n= 72. The correlation is summmarised in table below

Table 15

Group 1	Group 2	Degree of Freedom	T statistic	Correlation	95 % C.I.	P value
RESECTION_TIME_MIN	BLOOD_LOSS_DURING_PROCEDURE_LTR	50	-2.82	-0.37	-0.58–0.11	0.01

Table 16 Table with summary statistics of RESECTION TIME AND BLOOD LOSS

variable	n	Mean	SD	Median	Minimum	Maximum
BLOOD_LOSS_DURING_PROCEDURE_LTR	52	0.388	0.175	0.364	0.17	0.878
RESECTION_TIME_MIN	52	92.942	10.623	91.500	70.00	116.000

BLOOD LOSS BETWEEN CATHETRISED AND NON CATHETERISED PATIENTS

Figure 10 Boxplot Of Distribution Of CATHETARISATION DURATION in our Population

n this Figure we see Box plot of BLOOD_LOSS_DURING_PROCEDURE_LTR in 2 sub-groups of CATHETERISATION respectively .The individual jittered data points of BLOOD_LOSS_DURING_PROCEDURE_LTR are overlaid over transparent Boxplot for better visualisation. We see distribution of data in individual sub-groups of CATHETERISED_OR_NOT based on these box-plots. The lower edge of box plot represents -first quartile (Q1), Horizontal bar represents the median, Upper edge represnts third quartile (Q3), Two black lines (whiskers) emanating from box-plots signify range of non-outlier data for the particular sub-group. Lower whisker represents minimum(Q1- 1.5 interquartile range) non-outlier limit of BLOOD_LOSS_DURING_PROCEDURE_LTR and upper whisker represnts maximum(Q1+1.5interquartile range) of BLOOD_LOSS_DURING_PROCEDURE_LTR .Any data beyond whiskers of box-plots represents outliers in the sub-groups The big brown point in the box-plots represents mean BLOOD_LOSS_DURING_PROCEDURE_LTR of 2 groups and it has been annotated in the figure itself Summary Statistics of the groups is presented in table below

Table 17 Summary Table Of Catheterisation Duration within Groups

Group	n	Mean	SD	Median	Minimum	Maximum
Catheterised	32	0.396	0.163	0.37	0.170	0.784
Not Catheterised	20	0.375	0.197	0.35	0.174	0.878

The mean Blood loss in catheterised [ 0.4 ± 0.17 ] was non-significantly higher than Catheterised group [ 0.38 ± 0.18 ] . The mean difference was 0.01 and 95 % confidence interval of the difference was ( -0.07 - 0.1 ) . The p value was 0.76 . The t statistic was 0.3 and degree of freedom of the Welch unpaired two-sample t test was 51.22 .In Formal statistical notation this result is expressed as : t(51.22) = 0.3, p= 0.76

Effect On Qmax

Figure 3 Trace-Plot Of variation QMAX P with Time in BIPOLAR AND THULEP groups

This is a trace-plot of variation in Q MAX (on Y axis) with passage of of time on X-axis , Q MAX readings of BIPOLAR and THULEP TURP across various stages are connected by distinct colored trace-line . Individual readings have been plotted on graph and points have been jittered for better clarity . The raw data points representing two groups have different shapes as indicated in the legend. The colored points across the trace represent the mean readings in two groups and whiskers represent Standard error of measurements.

we can clearly see a positive trend/slope indicating a rise in Q MAX with passage of Time. T as we can see similar slopes of variation with Time between BIPOLAR and THULEP groups indicating average difference in Q MAX across groups doesnt vary with Time. . we decided to explore this intuitive graphical relationship with formal statistical tests.

Two way repeated measures ANOVA was done to assess difference between average value of QMAX at different time intervals and if there was inter-group differences between BIPOLAR AND THULEP . An interaction test was also conducted to see if inter-group variation of value within QMAX was affected by PROCEDURE_PERFORMED. Since Sphericity (homogeneity of variance between pairs of intra-group comparisons) assumption was not met.Greenhouse-Geisser Correction was applied . Inter group comparison of average difference between ** BIPOLAR AND THULEP** was non-significant with a p value of 0.1200 . In Formal statistical notation it is expressed as F(1, 44) = 2.49,p= 0.1200. Intra group comparison of average difference between value of Q MAX WITH TIME ( Evident in rise post procedure) was significant with a p value of 0.0001 . In Formal statistical notation it is expressed as F(4, 176) = 484.93,p= 0.0001. Interaction of value within variable with PROCEDURE_PERFORMED was non-significant with a p value of 0.4900 . In Formal statistical notation it is expressed as F(4, 176) = 0.85,p= 0.4900

Table 18. Average Q MAX, standard error and 95% Confidence limits in DRUG Groups

PROCEDURE_PERFORMED	Mean	Standard_Error	lower.CL	upper.CL
BIPOLAR	15.114	0.281	14.548	15.680
THULEP	15.741	0.281	15.175	16.307

Table 19. Average Contrast in Q MAX within DRUG Groups

contrast	estimate	SE	df	t.ratio	p.value
BIPOLAR - THULEP	-0.627	0.397	44	-1.577	0.122

We also wanted to look for a trend in our Q MAX variation group since it had a time varying component .We found a significant positive linear trend in intra-group difference with variable , 26.91+-0.67, p value= <0.001.It was qualified by a significant quadratic trend in intra-group difference with variable , -13.64+-0.79, p value= <0.001.

Table 20. Average QMAX variation with time

variable	Mean	Standard_Error	lower.CL	upper.CL
Q_MAX	8.162	0.274	7.621	8.703
Q_MAX_POD7	13.502	0.274	12.961	14.044
Q_MAX_POD_1_MONTH	17.609	0.274	17.068	18.150
Q_MAX_POD_3_MONTH	18.989	0.274	18.448	19.530
Q_MAX_POD_6_MONTH	18.874	0.274	18.333	19.416

Table 21. Average Contrast in QMAX across al Time periods

contrast	estimate	SE	df	t.ratio	p.value
Q_MAX - Q_MAX_POD7	-5.341	0.297	176	-17.953	0.001
Q_MAX - Q_MAX_POD_1_MONTH	-9.447	0.297	176	-31.759	0.001
Q_MAX - Q_MAX_POD_3_MONTH	-10.828	0.297	176	-36.398	0.001
Q_MAX - Q_MAX_POD_6_MONTH	-10.713	0.297	176	-36.012	0.001
Q_MAX_POD7 - Q_MAX_POD_1_MONTH	-4.107	0.297	176	-13.806	0.001
Q_MAX_POD7 - Q_MAX_POD_3_MONTH	-5.487	0.297	176	-18.445	0.001
Q_MAX_POD7 - Q_MAX_POD_6_MONTH	-5.372	0.297	176	-18.059	0.001
Q_MAX_POD_1_MONTH - Q_MAX_POD_3_MONTH	-1.380	0.297	176	-4.639	0.001
Q_MAX_POD_1_MONTH - Q_MAX_POD_6_MONTH	-1.265	0.297	176	-4.253	0.001
Q_MAX_POD_3_MONTH - Q_MAX_POD_6_MONTH	0.115	0.297	176	0.386	0.995

we see there is no variation across 3 months and 6 months rest are significant.

Effect On IPSS

Figure Trace-Plot Of variation IPSS P with Time in BIPOLAR AND THULEP groups

This is a trace-plot of variation in IPSS (on Y axis) with passage of of time on X-axis , IPSS readings of BIPOLAR and THULEP TURP across various stages are connected by distinct colored trace-line . Individual readings have been plotted on graph and points have been jittered for better clarity . The raw data points representing two groups have different shapes as indicated in the legend. The colored points across the trace represent the mean readings in two groups and whiskers represent Standard error of measurements we can clearly see a positive trend/slope indicating a fall in IPSS with passage of Time. T as we can see similar slopes of variation with Time between BIPOLAR and THULEP groups indicating average difference in Q MAX across groups doesnt varies with Time. . we decided to explore this intuitive graphical relationship with formal statistical tests.

Two way repeated measures ANOVA was done to assess difference between average value of IPSS and if there was inter-group differences between PROCEDURE_PERFORMED . An interaction test was also conducted to see if inter-group variation of value within variable was affected by PROCEDURE_PERFORMED. Since Sphericity (homogeneity of variance between pairs of intra-group comparisons) assumption was not met.Greenhouse-Geisser Correction was applied Inter group comparison of average difference between IPSSS between PROCEDURE_PERFORMED was significant with a p value of 0.0200 . In Formal statistical notation it is expressed as F(1, 46) = 5.38,p= 0.0200. Intra group comparison of average difference between value of variable was significant with a p value of 0.0001 . In Formal statistical notation it is expressed as F(4, 184) = 1230.78,p= 0.0001. Interaction of value within variable with PROCEDURE_PERFORMED was non-significant with a p value of 0.0500 . In Formal statistical notation it is expressed as F(4, 184) = 2.4,p= 0.0500

Since Intra group average differences of value within variable group was significant. We performed a test of contrasts. The Inter-group difference between BIPOLAR - THULEP was significant with mean difference of 0.97+-0.42, pvalue= 0.025

Table 22. Average IPSS, standard error and 95% Confidence limits in DRUG Groups

PROCEDURE_PERFORMED	Mean	Standard_Error	lower.CL	upper.CL
BIPOLAR	9.420	0.297	8.823	10.018
THULEP	8.446	0.297	7.849	9.044

Table 23. Average Contrast in IPSS between Groups

contrast	estimate	SE	df	t.ratio	p.value
BIPOLAR - THULEP	0.974	0.42	46	2.32	0.025

We also wanted to look for a trend in IPSS since it had a time varying component .We found a significant negative linear trend in intra-group difference with variable , -46.69+-0.77, p value= <0.001. It was qualified by significant quadratic n cubic trend as seen in parabolic trajectory of fall ofIPSS with time.

Table 24. Average IPSS variation with time

variable	Mean	Standard_Error	lower.CL	upper.CL
PREOP_IPSS	23.715	0.303	23.117	24.314
IPSS_ON_POD7	8.864	0.303	8.265	9.462
IPSS_ON_POD_1_MONTH	5.559	0.303	4.960	6.157
IPSS_ON_POD_3_MONTH	3.455	0.303	2.857	4.054
IPSS_ON_POD_6_MONTH	3.074	0.303	2.476	3.673

Table 25. Average Contrast in IPSS across al Time periods

contrast	estimate	SE	df	t.ratio	p.value
PREOP_IPSS - IPSS_ON_POD7	14.851	0.346	184	42.956	0.001
PREOP_IPSS - IPSS_ON_POD_1_MONTH	18.157	0.346	184	52.516	0.001
PREOP_IPSS - IPSS_ON_POD_3_MONTH	20.260	0.346	184	58.600	0.001
PREOP_IPSS - IPSS_ON_POD_6_MONTH	20.641	0.346	184	59.702	0.001
IPSS_ON_POD7 - IPSS_ON_POD_1_MONTH	3.305	0.346	184	9.560	0.001
IPSS_ON_POD7 - IPSS_ON_POD_3_MONTH	5.409	0.346	184	15.644	0.001
IPSS_ON_POD7 - IPSS_ON_POD_6_MONTH	5.790	0.346	184	16.746	0.001
IPSS_ON_POD_1_MONTH - IPSS_ON_POD_3_MONTH	2.103	0.346	184	6.084	0.001
IPSS_ON_POD_1_MONTH - IPSS_ON_POD_6_MONTH	2.484	0.346	184	7.186	0.001
IPSS_ON_POD_3_MONTH - IPSS_ON_POD_6_MONTH	0.381	0.346	184	1.102	0.806

we see there is no variation across 3 months and 6 months rest are significant.

Effect On PVR

Figure Trace-Plot Of variation in PVR with Time in BIPOLAR AND THULEP groups

This is a trace-plot of variation in PVR (on Y axis) with passage of of time on X-axis , PVR readings of BIPOLAR and THULEP TURP across various stages are connected by distinct colored trace-line . Individual readings have been plotted on graph and points have been jittered for better clarity . The raw data points representing two groups have different shapes as indicated in the legend. The colored points across the trace represent the mean readings in two groups and whiskers represent Standard error of measurements we can clearly see a positive trend/slope indicating a fall in PVR with passage of Time. T as we can see similar slopes of variation with Time between BIPOLAR and THULEP groups indicating average difference in PVR across groups doesnt varies with Time. . we decided to explore this intuitive graphical relationship with formal statistical tests.

wo way repeated measures ANOVA was done to assess difference between PVR across time and treatments. An interaction test was also conducted to see if inter-group variation of value within variable was affected by PROCEDURE_PERFORMED. Since Sphericity (homogeneity of variance between pairs of intra-group comparisons) assumption was not met.Greenhouse-Geisser Correction was applied Inter group comparison of average difference between PVR of 2 procedures was significant with a p value of 0.0300 . In Formal statistical notation it is expressed as F(1, 44) = 5.16,p= 0.0300. Intra group comparison of average difference between value of variable was significant with a p value of 0.0001 . In Formal statistical notation it is expressed as F(4, 176) = 230.85,p= 0.0001. Interaction of PVR with PROCEDURE_PERFORMED was non-significant (similar slopes) with a p value of 0.1200 . In Formal statistical notation it is expressed as F(4, 176) = 1.88,p= 0.1200

Since Intra group average differences of value within variable group was significant. We performed a test of contrasts. The Inter-group difference between BIPOLAR - THULEP was significant with mean difference of 4.8+-2.11, pvalue= 0.028 . Thus BIPOLAR TURP had a minimal but significant higher PVR than THULEP.

Table 26. Average PVR, standard error and 95% Confidence limits across treatments

PROCEDURE_PERFORMED	Mean	Standard_Error	lower.CL	upper.CL
BIPOLAR	47.878	1.495	44.866	50.89
THULEP	43.078	1.495	40.066	46.09

Table 27. Average Contrast in PVR between Groups

contrast	estimate	SE	df	t.ratio	p.value
BIPOLAR - THULEP	4.8	2.114	44	2.271	0.028

We also wanted to look for a trend in PVR since it had a time varying component . .We found a significant negative linear trend in PVR , -188.07+-7.6, p value= <0.001.qualified by significant quadratic n cubic trend.

Table 28. Average PVR variation with time

variable	Mean	Standard_Error	lower.CL	upper.CL
PVR	108.935	2.395	104.215	113.655
PVR_POD7	44.565	2.395	39.845	49.285
PVR_POD_1_MONTH	24.435	2.395	19.715	29.155
PVR_POD_3_MONTH	24.543	2.395	19.823	29.263
PVR_POD_6_MONTH	24.913	2.395	20.193	29.633

Table 29. Average Contrast in PVR across al Time periods

contrast	estimate	SE	df	t.ratio	p.value
PVR - PVR_POD7	64.370	3.398	176	18.942	0.001
PVR - PVR_POD_1_MONTH	84.500	3.398	176	24.866	0.001
PVR - PVR_POD_3_MONTH	84.391	3.398	176	24.834	0.001
PVR - PVR_POD_6_MONTH	84.022	3.398	176	24.726	0.001
PVR_POD7 - PVR_POD_1_MONTH	20.130	3.398	176	5.924	0.001
PVR_POD7 - PVR_POD_3_MONTH	20.022	3.398	176	5.892	0.001
PVR_POD7 - PVR_POD_6_MONTH	19.652	3.398	176	5.783	0.001
PVR_POD_1_MONTH - PVR_POD_3_MONTH	-0.109	3.398	176	-0.032	1
PVR_POD_1_MONTH - PVR_POD_6_MONTH	-0.478	3.398	176	-0.141	1
PVR_POD_3_MONTH - PVR_POD_6_MONTH	-0.370	3.398	176	-0.109	1

we see there is no variation across 1 months an, 3 month and 6 months ,rest are significant.

Effect On Prostate size

Figure Trace-Plot Of variation in PVR with Time in BIPOLAR AND THULEP groups

This is a trace-plot of variation in Prostate size (assessed by TAS) (on Y axis) with passage of of time on X-axis , Prostate size readings of BIPOLAR and THULEP TURP patients across various stages are connected by distinct colored trace-line . Individual readings have been plotted on graph and points have been jittered for better clarity . The raw data points representing two groups have different shapes as indicated in the legend. The colored points across the trace represent the mean readings in two groups and whiskers represent Standard error of measurements we can clearly see a negative trend/slope indicating a fall in Prostate size with passage of Time initially followed by alate surge at 6 months.. as we can see different slopes of variation with Time between BIPOLAR and THULEP groups indicating average difference in PVR across groups varies with Time(Thulep had a sharper fall at 3 months). . we decided to explore this intuitive graphical relationship with formal statistical tests.

Two way repeated measures ANOVA was done to assess difference between average Prostate size and if there was inter-group differences between PROCEDURE_PERFORMED . An interaction test was also conducted to see if inter-group variation of value within variable was affected by PROCEDURE_PERFORMED. Since Sphericity (homogeneity of variance between pairs of intra-group comparisons) assumption was not met.Greenhouse-Geisser Correction was applied . Inter group comparison of average difference between Prostate size of PROCEDURE_PERFORMED was significant with a p value of 0.0030 . In Formal statistical notation it is expressed as F(1, 69) = 9.57,p= 0.0030. Intra group variation of Prostate size with time was significant with a p value of 0.0001 . In Formal statistical notation it is expressed as F(2, 138) = 687.81,p= 0.0001. Interaction of Prostate size with traetment modalities was significant with a p value of 0.0001 . In Formal statistical notation it is expressed as F(2, 138) = 19.91,p= 0.0001

Since Intra group average differences of value within variable group was significant. We performed a test of contrasts. The Inter-group difference between BIPOLAR - THULEP was significant with mean difference of 7.52+-2.43, pvalue= 0.003 implying on an average BIPOLAR group had about 7 gm higher prostate size.

Table 30. Average Prostate size , standard error and 95% Confidence limits across treatments

PROCEDURE_PERFORMED	Mean	Standard_Error	lower.CL	upper.CL
BIPOLAR	58.994	1.787	55.430	62.559
THULEP	51.907	1.787	48.342	55.472

Table 31. Average Contrast in Prostate size between Groups

contrast	estimate	SE	df	t.ratio	p.value
BIPOLAR - THULEP	7.087	2.527	69	2.804	0.007

we also wanted to look for a trend in our variable group since it had a time varying component .We found a significant negative linear trend in intra-group difference with variable , -33.54+-1.09, p value= <0.001.We also found a significant positive quadratic trend (indicating alate rise from 3 months to 6 months) in intra-group difference with variable , 39.05+-1.89, p value= <0.001

Table 32, Average Prostate size variation with time

variable	Mean	Standard_Error	lower.CL	upper.CL
PROSTATE_SIZE	79.081	1.399	76.306	81.856
REDIUAL_PROSTATE_AT_3_MONTH_CC	42.782	1.399	40.007	45.557
REDIUAL_PROSTATE_AT_6_MONTH_CC	44.489	1.399	41.715	47.264

Table 33. Average Contrast in Prostate size across al Time periods

contrast	estimate	SE	df	t.ratio	p.value
PROSTATE_SIZE - REDIUAL_PROSTATE_AT_3_MONTH_CC	36.299	1.039	138	34.930	0.001
PROSTATE_SIZE - REDIUAL_PROSTATE_AT_6_MONTH_CC	34.592	1.039	138	33.287	0.001
REDIUAL_PROSTATE_AT_3_MONTH_CC - REDIUAL_PROSTATE_AT_6_MONTH_CC	-1.707	1.039	138	-1.643	0.231

we see all three contrasts between prostate sizes are significant indicating a constant change(initially larger fall n later a small rise)