Bi l i b ildiBio-aerosols in building drainage and plumbing

tsystems: cross contamination, monitoring

and prevention.

Dr Michael GormleyDrainage Research GroupSchool of the Built Environment Heriot-Watt UniversityEdinburgh

ContentsContents

The building drainage system as The building drainage system as a bioaerosol transmission route

Transmission study of the drainage system of a hospitaldrainage system of a hospital

Monitoring method for minimising Monitoring method for minimisingbioaerosol transmission from the drainage systemdrainage system

IntroductionBio-aerosols- what are they?y

Bioaerosols are defined as airborne particles, largemolecules or volatile compounds that are living, containliving organisms or were released from livingliving organisms or were released from livingorganisms. The size of a bioaerosol particle may varyfrom 100 microns to 0.01 micron. The behaviour ofbioaerosols is governed by the principles of gravitation,electromagnetism, turbulence and diffusion.

IntroductionRelative size of particles

100 microns

10 microns1 micron

0.01 microns

This is a scale representation of the relative size of pollen, pollen spores, bacteria and viruses. The scale of this diagram is roughly 8000:1. Each of the dots on this screen version represent 15 viruses or virions In this diagramdots on this screen version represent 15 viruses, or virions. In this diagram, approximately 100,000 of these virions fit within the 100 micron circle representing the pollen. In actuality, many millions of virions could fit within the cross-section of a pollen.

Bio-aerosol generation and detection

Bioaerosols, particularly those containing viruses are particularly difficult to isolate and identify.

This task is made even more difficult due to the unsteady nature of flows in building d i tdrainage systems.

Building drainage system: mechanisms for air flow and pressure transient generation

direction of air entrainmentflow and pressure transient generation

DischargingIncreased entrainedDischarging

appliance increases entrained airflow and generates

ti t i t

entrainedairflow

negative transients.

Trap seal depletion

Negative transient

depletionIncreased water

flow.

Building drainage system: mechanisms for air flow and pressure transient generationflow and pressure transient generation

operating appliance

Positive transient affects Trap seal

all trapsTrap seal

deflections

Airflow reduced due to stack base

surchargesurcharge

Pressure transients in system can cause traps to blow out-

http://www.youtube.com/watch?feature=player detailpage&v=d vNL

ture player_detailpage&v d_vNLMCZ9jQ

video

The attached video is anThe attached video is an extreme example – but it is real – most common symptom of smallersymptom of smaller pressure transients is bubbling through a trap, you may have seen this in y ya toilet bowl.

Airflow and pressure transient modelling- AIRNET and current limitations

A method of characteristics based numerical model.

Finite difference scheme

Developed and validated over 30 years at Heriot- Watt University – initiated by, and

ti t b i i d b th k f J hcontinues to be inspired by, the work of John Swaffield.

Building Drainage System modelling in AIRNET (3)

When

P 

Δt =    Δx  

Δx  A′ B′

C‐C+

        (u+c)max  

A

R  S BC 

0||42

tuufccuu

For C+ - Line PR

02

||41

D

uufccuu RRRRPRP

dx u c

when

dtu c

02

||41

2

Dtuufccuu SSSSPSP

For C- - Line PS Building drainage system boundary conditions

cudtdx

when

AIRNET modelling

Modelling of complex networks such as the O2 DomeThi i th fi t l d d i tThis is the first sealed drainage system ever constructed. It has no penetrations through the roof. As drainage systems go, it is unique.The approach designed by M.Gormley and

50 Storey housing block in Hong KongModelling led to novel approaches to preventingpp g y y

J.A.Swaffield from Heriot-Wattapproaches to preventing excessive positive pressures using P.A.P.A.TM

modelling air pressure and flow in large systems

St kSt k

W.c.

Roof lineStack termination

W.c.

Roof lineStack termination

Roof lineStack terminationRoof lineStack termination

Entrained air

W.c.

W.c.

‘Wet’ stackParallel vent

W.c.

W.c.

‘Wet’ stackParallel vent

W.c.

W.c.

W.c.

W.c.

W.c.

Roof lineStack termination

W.c.

Roof lineStack termination

air

Sewer connection

Wet stackParallel vent stack with cross connections Sewer

connection

Wet stackParallel vent stack with cross connections

W.c.

Sewer

‘Wet’ stackParallel vent stack with cross connections

W.c.

Sewer

‘Wet’ stackParallel vent stack with cross connectionsW c

W.c.

W c

W.c.W.c.

Roof lineStack termination

W.c.

Roof lineStack termination

connectionconnectionW.c.

Sewer connection

‘Wet’ stackParallel vent stack with cross connections

W.c.

Sewer connection

‘Wet’ stackParallel vent stack with cross connections

W.c.

W.c.

‘ ’

W.c.

W.c.

‘ ’

Combined waste fluids and waste in appliance

Sewer connection

‘Wet’ stackParallel vent stack with cross connections Sewer

connection

‘Wet’ stackParallel vent stack with cross connections

in appliance water flows

St kSt k

W.c.

Roof lineStack termination

W.c.

Roof lineStack termination

Roof lineStack terminationRoof lineStack termination

Potential contaminated air ingress if trap seal lost

W.c.

W.c.

‘Wet’ stackParallel vent

W.c.

W.c.

‘Wet’ stackParallel vent

W.c.

W.c.

W.c.

W.c.

W.c.

Roof lineStack termination

W.c.

Roof lineStack termination

p

Sewer connection

Wet stackParallel vent stack with cross connections Sewer

connection

Wet stackParallel vent stack with cross connections

W.c.

Sewer

‘Wet’ stackParallel vent stack with cross connections

W.c.

Sewer

‘Wet’ stackParallel vent stack with cross connectionsW c

W.c.

W c

W.c.W.c.

Roof lineStack termination

W.c.

Roof lineStack termination

connectionconnectionW.c.

Sewer connection

‘Wet’ stackParallel vent stack with cross connections

W.c.

Sewer connection

‘Wet’ stackParallel vent stack with cross connections

W.c.

W.c.

W.c.

W.c.

Airflow may be drawn from any of the connected sub – section drainage

Sewer connection

‘Wet’ stackParallel vent stack with cross connections Sewer

connection

‘Wet’ stackParallel vent stack with cross connections

sub section drainage systems – total interconnection.

Limitations

• Calculations do not include important pbioaerosol fluid dynamics such as;• Brownian Motion• Gravitation• Electrical Forces• Thermal Gradients & Electromagnetic Radiation• Turbulent Diffusion• Inertial Impaction• Inertial Impaction• Particle Shape

However• Flow direction and rate can be calculated –

approximations of likely bio-aerosol transport mechanisms can be made.

Modelling flow rate and direction

6

0

2

4

6om

, - d

own

stac

k,

nd

-8

-6

-4

-2

ned

airf

low

, + in

to r

oolit

res/

seco

n Modelling confirms the establishment of an air exchange between the bathroom and the vertical

-12

-10

0 5 10 15 20

Time, seconds.Entr

ain

The building drainage systemInterconnection- all parts of the building are interconnectedinterconnected

3F

4F

2F

3F

Waste water from other hospital 

b ildi i GF

1F

buildings i.e. labs, morgue.

GF

To main sewer

The building drainage systemSARS OutbreakSARS Outbreak

Press Release WHO/70

“droplets originating from virus

26 September 2003:

droplets originating from virus-rich excreta…re-entered into

residents apartments via sewage and drainage systems where there were strong upwards air flows, inadequate ‘traps’ andflows, inadequate traps and non-functional water seals.”

SARS OutbreakTransmission routeTransmission route

Bioaerosols carried to adjacent buildings

by wind current

Bioaerosols transmittedBioaerosols transmitted to adjacent apartment

Infected person introduces virus to drainage system

Bioaerosols formed as waste is flushedas waste is flushed

The building drainage systemNew threatsNew threats

Airborne transmission evidenceForgotten knowledgeForgotten knowledge

1907: cultured airborne Serratia marcescens (then termed Bacillus prodigiosus) from drainage

systems and detected airborne transport from one hospital building to another via the sewer drain.

Sir William Heaton Horrocks(1859-1941)

Horrocks – In good companyThe Royal SocietyThe Royal Society

I N tIsaac Newton

Charles Babbage

James Watt

Horrocks – Other Successes

C fi d th t th f ‘M lt F ’• Confirmed that the cause of ‘Malta Fever’ was bacteria passed on through goats milk.

• Developed methods for testing and purifying drinking water

• Published book on bacteriology of water, one of the first of its kindof the first of its kind.

Horrocks, William Heaton (1901). An Introduction to the Bacteriological E amination of Waterto the Bacteriological Examination of Water. London: J. & A. Churchill.

Pathogen transmission studyHospital buildingHospital building

Environmental conditions

NorovirusOutbreak

3F

4F

2F

3F

Bioaerosoltransmission

Waste water from other hospital 

b ildi i GF

1F

buildings i.e. labs, morgue.

GF

To main sewer Wastewater

contamination

Pathogen transmission studyHospital buildingHospital building

Air sampling

Pitot tube measuring downward airflow

Collection swab• Isolation of bioaerosols using collection swab (UTM-RT)

Static pressure tube • Temperature and humidity within

drainage stack (USB data logger)

• Air flow and direction (pitot tube)

Pitot tube measuring upward airflow

USB temperature & relative humidity data loggerWastewater sampling

• Collection of wastewater from main underground drain

Pathogen transmission studyReal Time Polymerase Chain ReactionReal Time Polymerase Chain Reaction

Samples extracted using NucliSens® easyMAG™ system

Ct ≤ 29 St iti tiCt ≤ 29 Strong positive reaction (abundant target nucleic acid)

Ct 30-37Positive reaction (moderate amount of target nucleic acid)

Amplification, detection and analysis performed in an ABI 7500 RT-PCR system

( g )Ct 38-40Weak reaction

(minimal amounts of target nucleic acid)

Pathogen transmission studyRT-PCR ResultsRT PCR Results

T t d tTest date Norovirus GI Norovirus GIISewer Stack 1 Stack 2 Stack 3 Sewer Stack 1 Stack 2 Stack 3

01/03/2011 U U U U U U U U10/03/2011 U U U U 2 U U U10/03/2011 U U U U 25 U U U16/03/2011 U U U U 25 U U U23/03/2011 U U U U 35 U U U30/03/2011 U U U U 40 U U U30/03/2011 U U U U 40 U U U05/04/2011* U U U U 37 U U U26/05/2011 N/A U U U N/A U U U

U UndetectedCt ≤ 29 Strong positive reaction (abundant target nucleic acid)Ct 30-37 Positive reaction (moderate amount of target nucleic acid)Ct 38 40 Weak reaction (minimal amounts of target nucleic acid)Ct 38-40 Weak reaction (minimal amounts of target nucleic acid)

*a swab of the inside surface of Stack 1 taken on this date also returned undetected for all tests

Pathogen transmission studyRT-PCR ResultsRT PCR Results

Samples were also tested for Clostridium deficile but was undetected.This was due to the fact that Cdiff produces spores which are not amenable to man of the PCR assa s a ailablemany of the PCR assays available.

CYCLE NUMBER AMOUNT OF DNA0 11 22 4

1400000000

1600000000

2 43 84 165 326 647 128

400000000

600000000

800000000

1000000000

1200000000

AM

OU

NT

OF

DN

A8 2569 512

10 1,02411 2,04812 4,09613 8 192

0

200000000

0 5 10 15 20 25 30 35

PCR CYCLE NUMBER

A13 8,19214 16,38415 32,76816 65,53617 131,07218 262,144

1000001000000

10000000100000000

100000000010000000000

T O

F D

NA

18 262,14419 524,28820 1,048,57621 2,097,15222 4,194,30423 8,388,608

110

1001000

10000

0 5 10 15 20 25 30 35

AM

OU

NT24 16,777,216

25 33,554,43226 67,108,86427 134,217,72828 268,435,45629 536 870 912

31

0 5 10 15 20 25 30 35

PCR CYCLE NUMBER29 536,870,91230 1,073,741,82431 1,400,000,00032 1,500,000,00033 1,550,000,00034 1,580,000,000

Pathogen transmission studyTemperature and Humidity

98

100

%)

Temperature and Humidity

94

96

Hum

idity

(% Average humidity = 96.6%

90

92

23/03/11 24/03/11 25/03/11 26/03/11 27/03/11 28/03/11 29/03/11 30/03/11 31/03/11D tDate

26

28

30

re (o

C ) Average temperature = 24.3%

22

24

26

Tem

pera

tu

2023/03/11 24/03/11 25/03/11 26/03/11 27/03/11 28/03/11 29/03/11 30/03/11 31/03/11

Date

Pathogen transmission studyAirflow results – this proves that the interconnection hypothesis is validvalid

30 Airflow Up

20Airflow Down

0

10

rate (l/s)

‐10

00 1 2 3 4 5 6

Airflow r

‐20

‐30Time (minutes)

Additonal Domestic system testsDrainage System Schematic

Roof Level (height above collection drain = 10 m

Floor 3

Open pipe for testing (WC Removed) Simulates open trap Anemometer located here.

Bath & sink

Floor 2

Kitchen WC

To main sewer

Floor 1

2 x Bathrooms Utility room 2 x WCs

Smoke Pellets To main sewer inserted here

Collection drain

20

Air Flowrate

14

16

18

Note: Similar

10

12

ow ra

te (l/s)

AirFlow

Similar Airflows to those

4

6

8Airfl AirFlow

recorded in Amoy Gardens

0

2

4

0 500 1000 1500 2000 2500 3000 3500 40000 500 1000 1500 2000 2500 3000 3500 4000Time (secs)

The DYTEQTA SystemAutomated monitoring methodAutomated monitoring method

Defective fixture trap seals increase risk of bioaerosol transmission via the building drainage network

Dyteqta is a sonar-like method for establishing the status of each fixture trap seal in a building

Based on reflected wave theory

Using a sinusoidal air pressure a e ens res the test is nonwave ensures the test is non-

destructive

System validated by: modelling System validated by: modelling, laboratory investigations and extensive site testing

Case study BuildingsCase study Buildings

The DYTEQTA SystemCase studies

1

60

80

Case studiesTime change detection algorithm

0.5

20

40

r gau

ge)

0

-20

0

20

0.00 0.05 0.10 0.15 0.20 0.25 0.30 Dt >

h

re (m

m w

ater

-0.5

60

-40

-20

Pre

ssur

Defect free baseline

-1-80

-60

Time (seconds)

Test pressure responseDt > h

Is Dt > h over calibration period? NO, trace is reliable.Is Dt > h during test period? YES, at tD = 0.066 seconds.Depleted trap location? T12.

The DYTEQTA SystemCase studiesCase studies

60D d (AIRNET)

50

n (m

)

Dundee (AIRNET)Dundee (PROBE)Arrol (AIRNET)Arrol (PROBE)

30

40

ap lo

catio

n Arrol (PROBE)Glasgow (AIRNET)Glasgow (PROBE)

20

edic

ted

tra

0

10Pre

00 10 20 30 40 50 60

True trap location (m)

The building drainage systemTransmission of bioaerosolsTransmission of bioaerosols The building drainage system

interconnects all parts of a buildingg

Potential cross-transmission route for bio-aerosols.

Every building tested had empty water trap seals.

Healthcare building drains have a distinctly ‘hospital smell’ they do not y ynecessarily smell malodorous.

Norovirus GII isolated from t t l d f i d iwastewater sampled from main drain

of a hospital building, confirming contamination during an outbreak.

The building drainage systemTransmission of bioaerosolsTransmission of bioaerosols

Environmental conditions within theEnvironmental conditions within the drainage system are conducive to bio-aerosol circulation

Current work underway to replicate the Horrocks work reported in the Royal Society proceedings in 1907Royal Society proceedings in 1907 and extend the investigations on the identification of specific pathogens in airflows.

This work has confirmed that bacteria such as psuedomonas spp. Can be carried on airstreams inside a building drainage system.

Th k f li t iThank you for listening