19-A reviewCOVID-final_3-PRIMARY.qxd 9/7/20 8:30 PM Page 574

REVIEW ARTICLE

A review of smell and dysfunction in COVID-19 patients

Danny Kit Chung Wong 1, Hardip Singh Gendeh 2, How Kit Thong 1, Sai Guan Lum 2, Balwant Singh Gendeh 3, Aminuddin Saim 1, Salina Husain 2

1Department of -Head and Neck Surgery, Faculty of Medicine, KPJ Healthcare University College, Negeri Sembilan, Malaysia, 2Department of Otorhinolaryngology-Head and Neck Surgery, Faculty of Medicine, Universiti Kebangsaan Malaysia, Kuala Lumpur, Malaysia, 3Hospital Pantai Kuala Lumpur, Kuala Lumpur, Malaysia

ABSTRACT INTRODUCTION The severe acute respiratory syndrome coronavirus-2 (SARS- Introduction: Multiple anecdotal reports suggest that smell COV-2) or Coronavirus disease 2019 (COVID-19) has infected and taste loss were early subclinical symptoms of COVID-19 more than 5.2 million people with 330,000 over fatalities as patients. The objective of this review was to identify the of May 2020. 1 The outbreak was first detected in Wuhan, incidence of smell and taste dysfunction in COVID-19, China in December 2019 and has rapidly spread globally determine the onset of their symptoms and the risk factors over the last six months. It is transmitted via droplet of , , or for COVID-19 inhalation through the upper respiratory tract and was . declared a global pandemic by the World Health Organization (WHO) on 11th March 2020. Methods: We searched the PubMed and Google Scholar on 15th May 2020, with search terms including SARS-COV-2, The typical general symptoms of COVID-19 are fever, cough, coronavirus, COVID-19, hyposmia, anosmia, ageusia and and shortness of breath. Other symptoms include sore throat, dysgeusia. The articles included were cross sectional nasal congestion, body aches, myalgia, headaches, nausea studies, observational studies and retrospective or and vomiting, diarrhoea and abdominal . 2 Anosmia and prospective audits, letters to editor and short ageusia or smell and taste dysfunction are not listed communications that included a study of a cohort of symptoms as part of the screening criteria. patients. Case reports, case-series and interventional studies were excluded. Since being declared a pandemic, there have been anecdotal reports suggesting that smell and taste loss were early Discussion: A total of 16 studies were selected. Incidence of subclinical symptoms of COVID-19. Otorhinolaryngology smell and taste dysfunction was higher in Europe (34 to (ORL) or Ear, Nose and Throat (ENT) societies have alerted 86%), North America (19 to 71%) and the Middle East (36 to physicians to the risk of patients having COVID-19 with 98%) when compared to the Asian cohorts (11 to 15%) in isolated smell or taste dysfunction before the general COVID-19 positive patients. Incidence of smell and taste symptoms occur (i.e. fever, dry cough, fatigue and shortness dysfunction in COVID-19 negative patients was low in of breath). 3 comparison (12 to 27%). Total incidence of smell and taste dysfunction from COVID-19 positive and negative patients In mid-March 2020, the German virologist Professor Hendrick from seven studies was 20% and 10% respectively. Streeck, reported that more than two in three of 200 Symptoms may appear just before, concomitantly, or confirmed COVID-19 patients interviewed in a hospital immediately after the onset of the usual symptoms. Occurs reported loss of smell and taste. 4 France in a press release said predominantly in females. When occurring immediately after anosmia without nasal obstruction is a pathognomonic sign the onset of the usual symptoms, the median time of onset of COVID-19 infection after their interview with 55 COVID-19 was 3.3 to 4.4 days. Symptoms persist for a period of seven patients. 5 Iran showed a dramatic increase in complains of to 14 days. Patients with smell and taste dysfunction were anosmia across multiple regions since the start of the reported to have a six to ten-fold odds of having COVID-19. coronavirus outbreak. 6 Conclusion: Smell and taste dysfunction has a high Although some countries have passed their peak of incidence in Europe, North America, and the Middle East. , others such as Brazil, Russia, Peru and Mexico are The incidence was lower in the Asia region. It is a strong risk seeing an increase in cases. 7 According to Professor Claire factor for COVID-19. It may be the only symptom and should Hopkins, the President of the British Rhinological Society be added to the list of symptoms when screening for COVID- (BRS), in the UK, there were a significant number of essential 19. workers who had anosmia or hyposmia but did not meet criteria for a swab test and were forced to go to work. These KEYWORDS: COVID-19; SARS-COV-2; smell; taste; dysfunction people potentially had COVID-19 and their diagnosis could have been missed thus contributing to the spread of disease. 8

This article was accepted: 25 June 2020 Corresponding Author: Hardip Singh Gendeh Email: [email protected]

574 Med J Malaysia Vol 75 No 5 September 2020 19-A reviewCOVID-final_3-PRIMARY.qxd 9/7/20 8:30 PM Page 575

A review of smell and taste dysfunction in COVID-19 patients

As a result, this has led to the announcement of BRS on the In all, both searches revealed 630 results and a total of 16 21st March 2020 to call for healthcare professionals to don manuscripts were included in this narrative review tabulated adequate personal protective equipment (PPE) in view of in Table I. The studies comprised of 3 prospective audits, 5 strong evidence of anosmia being a valid symptom for retrospective audits, 6 short communications and 2 pre- COVID-19. 3 Meanwhile in the USA, the Center for Disease prints. 10-25 The studies were published between 25th February Control and Prevention (CDC) in April 2020 added loss of 2020 to 12th May 2020. taste or smell to its list less common symptoms that may arise two to 14 days after exposure to the COVID-19 virus. 9 Anatomy and Physiology of Olfaction Furthermore, many nations have yet to include smell and/or The zone of smell of the classical in the nose test disorders as a screening symptom for COVID-19. lies in the , located at the roof of the . It covers an area of about 5cm 2 on either side of There were three objectives with the present review. Firstly, it the nose, including superior turbinates, septum and ethmoid was to identify the incidence or frequency of smell and taste bone. The structure of the nose is designed, in part, to direct dysfunction in COVID-19; secondly, to determine the onset of inspired air toward the olfactory epithelium. The the smell and taste dysfunction whether it was before, during neuroepithelium is characterised by the presence of olfactory or after the general symptoms of COVID-19 (i.e., fever, cough, neurons whose axons project through the of shortness of breath) and finally, to determine if smell and the , and synapse with neurons in the central taste dysfunction is a significant risk factor for COVID-19. olfactory nervous system. The group of axons form an olfactory tract that connects to on the ventral surface of the . The axons then travel to the olfactory cortex in the inferior and medial . MATERIALS AND METHODS This literature review was narrative in nature. Latest Some project to limbic system and hypothalamus, where manuscripts within the past six months (coinciding with the smells associated with long-term memory and emotional start of COVID-19) on COVID-19 from peer-reviewed journals response. Newer studies revealed a more extensive were included. Searches was conducted using the PubMed distribution of olfactory epithelium that involves the and Google Scholar database on 15th May 2020. anterolateral part of middle turbinate and posterior nasal septum. 26 Search terms used were “SARS-CoV-2 AND hyposmia OR anosmia AND coronavirus” (Search 1) and “SARS-CoV-2 Odorant particles are inhaled through the nose and carried AND ageusia OR dysgeusia AND coronavirus” (Search 2). by the turbulent airflow to the olfactory epithelium superiorly Relevant references from individual articles were retrieved and dissolve into mucous membrane. These odorant and included. molecules are transported by binding proteins to the olfactory receptors within the cell membrane of an olfactory dendrite. The inclusion criteria were cross sectional studies, Binding of the odorants to the specific olfactory receptors observational studies and retrospective or prospective audits, induces graded membrane potential in the olfactory neurons, letters to editor and short communications that included a which transmits signals to the brain. study of a cohort of patients. Case reports, case-series, reviews, and interventional studies were excluded. Both Besides the normal route, some of the odorants reach the published and unpublished (pre-print) materials were olfactory epithelium by ascending through the nasopharynx included. and posterior choanae. This retronasal olfaction has contributed in part to the sensation of flavour during Data from shortlisted papers were divided equally between consumption of food and liquids. 26 This may explain the four researchers, two each for Search 1 and Search 2 reason why taste sensation can be affected when the olfactory respectively who obtained full texts and reviewed the function is impaired. manuscript. Risk bias was minimised by adhering to the objectives of this study. Duplicate articles were removed. A Smell and Taste Dysfunction In COVID-19 second author re-reviewed the full text manuscripts. Olfactory dysfunction presenting as anosmia or hyposmia Significant and relevant results were tabulated based on has an overall prevalence rate of 3 to 20%. It usually affects authors, study designs, sample sizes, symptoms, age groups, males predominantly and is worse in the older age groups. 27- gender, duration, and recommendations. Wherever possible 30 Olfactory dysfunction in the viral upper respiratory tract results collected on smell and taste dysfunction were infections is not uncommon and is present in up to 30% of tabulated based on COVID-19 positive or negative patients patients. 27 Causal viruses include rhinovirus, parainfluenza, with respective sample size. adenovirus, Epstein-Barr virus and coronavirus. 31

Additional analysis was performed to identify the incidence All reviewed studies involved positive or confirmed COVID-19 or frequency of smell and taste dysfunction in COVID-19. cases. Majority of studies had lab-based confirmation in the Only case control studies with COVID-19 negative patients as form of reverse transcriptase polymerase chain reaction (RT- a control were used for incidence calculation. The calculation PCR) of nasopharyngeal and/or throat swabs and Bronchial- of incidence is based on the total number of smell and taste Alveolar-Lavage (BAL). disorders in COVID-19 positive patients/total number of Covid-19 positive and negative patients that were included. There were a total of seven studies that looked into smell and taste dysfunction and compared its symptoms between COVID-19 positive and negative patients. 11,18-20,23-25 The total

Med J Malaysia Vol 75 No 5 September 2020 575 19-A reviewCOVID-final_3-PRIMARY.qxd 9/7/20 8:30 PM Page 576

Review article

smell and taste dysfunction complaints in COVID-19 positive (11 to 15%). Some authors believed that this was because patients was 5077, almost twice that of smell and taste there was a decreased focus on nasal obstruction, dysfunction in non-COVID-19 patients at 2648 (ratio 1.9:1). rhinorrhoea and chemosensory complaints in these cohorts There were 65.1% (5077/7798 patients) COVID-19 positive due to the initial clinical impact, rapid spread and severity of patients who had smell and taste dysfunction smell and taste the disease. 13,14 dysfunction compared to only 21.2% (2648/12491 patients) who were COVID-19 negative with smell and taste However, a large epidemiological report of 1099 patients dysfunction. from China documented that cough (68%), fever (44%), fatigue (34%), and sputum production (34%) were the most Therefore, the total incidence of smell and taste dysfunction prevalent symptoms. 2 Nasal congestion or rhinorrhoea from COVID-19 positive and negative from the seven studies accounted for 4-6% of their cohort that was similar to above is 30 in 100 patients (30%). The incidence of smell and another study in China. 2,33 Tonsil swelling and throat taste dysfunction among COVID-19 positive patients is 20 in congestion accounted for 2% of complains and sore throat 100 patients (20%). The incidence of smell and taste 14%. 2 dysfunction among COVID-19 negative patients is 10 in 100 patients (10%). Therefore it is possible that the clinical differences in patients are due to the affinity of the COVID-19 virus to different In our review, only two studies reviewed anosmia and tissue sites in different ethnic groups as alluded by Lechien et hyposmia alone. 19,23 The rest of the studies also reviewed al., or the possibility of a mutated viral genome in Europe ageusia and dysgeusia and documented a very close relation and North America. 13,14 between smell and taste dysfunction. They found that among the COVID-19 positive patients the incidence of smell and Pathophysiology of Anosmia taste dysfunctions ranged from 11% to 86%. 14,17 This was Initially, it was thought that mechanical obstruction due to higher compared to COVID-19 negative patients where the excess production of mucous during active infection was the incidence ranged from 12 to 27%. 11,18,20,24 cause of anosmia, however persistent anosmia in patients with normal acoustic rhinometry post infection suggests that Smell and taste dysfunction in COVID-19 appeared to have there may be a different pathophysiology resulting in more different characteristics from the other viruses described lasting damage. 31 above. They are often the first symptom to be reported in mildly symptomatic patients in COVID-19. 22 They seem to COVID-19 enters into its human host by a cell surface appear just before, concomitantly or immediately after the receptor known as angiotensin‐converting enzyme 2 (ACE2), onset of the general symptoms of fever, cough, myalgia and which is expressed in the human airway epithelia, lung shortness of breath. When appearing before the onset of the parenchyma, vascular endothelia, kidney cells, and small general symptoms, they can occur 11.8 to 58% of the intestine cells. 34 ACE2 receptor binds to the S1 spike time. 11,12,14,22,23 When occurring immediately after the onset of glycoprotein in the viral envelope. The virus then enters the the general symptoms, the median time of onset was 3.3 to host cell through endocytosis. 4.4 days. 13,21 The symptoms of anosmia, hyposmia, ageusia and dysgeusia usually persist for a period of about seven to However, it requires a second protein known as 14 days with a majority improving after two weeks. 11,13-16,25 transmembrane protease serine 2 (TMPRSS2). TMPRSS2 is a protease that resides in the endosomal compartment that They are associated with mild nasal obstruction 13 to 50% of primes and cleaves the S1 spike glycoprotein, allowing fusion the time and usually have no other rhinological symptoms of the viral envelope with the endosomal compartment of the (i.e., nasal congestion, rhinorrhoea, itchiness, facial pain) host cell. 35 Respiratory viruses such as influenza A, evident by their relatively low SNOT-22 score in the Spinato parainfluenza and Japanese encephalitis have known to use et al. cohort (i.e., median score of four). 11,12,20-23,25 In some the olfactory nerves to gain access to the central nervous studies, olfactory dysfunction was the only symptom system. 15,36 Similarly, SARS-coronavirus have shown reported. 14,21 Predominantly women and younger patients intracranial trans-neural access via olfactory bulb in mice presented with chemosensory dysfunction. 10-16,18-25 Patients models following intranasal inoculation. 37 Therefore, it is with smell and taste dysfunction were reported to have a six believed that SARS-CoV-2 may share the same properties of to ten fold odds of having COVID-19. 18,25 its predecessors.

There were two studies reporting patients with COVID-19 Studies from the 1960s have shown that coronovirus’s are with smell and taste dysfunction having able to differentiate neuroinvasive and neurotropic. That means that it can , saltiness, and bitterness, controlled by specific penetrate the central nervous system (CNS) and infect taste afferents ranging between 60 to 88.9%. This suggests neurons and glial cells and could induce an overreaction of that COVID-19 may have an affinity or tropism to specific the immune system. 38 olfactory sensors. 11,32 This has been seen in Wuhan, China where some patients Demographic Distribution of smell and taste dysfunction with COVID‐19 also showed neurologic signs, such as in COVID-19 headache, nausea, and vomiting. 34 A recent study found Then incidence of smell and taste dysfunction was higher in expression of both ACE2 and TMPRSS2 in nasal respiratory Europe (34 to 86%), North America (19 to 71%) and the tissue, as expected, but also in olfactory neuroepithelia, Middle East (36 to 98%) when compared to the Asian cohorts central nervous systems and lymphoid tissues. 39

576 Med J Malaysia Vol 75 No 5 September 2020 19-A reviewCOVID-final_3-PRIMARY.qxd 9/7/20 8:30 PM Page 577

A review of smell and taste dysfunction in COVID-19 patients

This was confirmed by another study that also found that AUTHORS' OPINIONS nasal epithelial cells, including clusters of goblet cells and Smell and taste dysfunction appear to be a common early ciliated cells, have the highest expression among all clinical presentation especially in Europe, North America, investigated cells in the respiratory tree. 40 Moreover, as and the Middle East. Olfactory dysfunction is highly COVID-19 is an enveloped virus, its release does not require prevalent and has a high sensitivity rate as an early marker cell lysis. Thus, the virus might exploit existing secretory for COVID-19. 11,13,18-25 In some cases anosmia or hyposmia pathways in nasal goblet cells for low-level, continuous- may be the first and only symptom. 14,21 release at the early stage with no overt pathology. 40 The symptoms of anosmia, hyposmia, ageusia and dysgeusia There are also suggestions that besides ACE2, difference in could be a useful, harmless, inexpensive and efficient way to levels of expression of spike S protein for viral entry as evaluate potential patients for COVID-19. This allows mentioned above and nucleoplasmid N protein for patients to be tested early for COVID-19 and self-isolate. It transcription may influence the behaviours of COVID-19 would also allow for healthcare workers to employ universal within different regions and ethnicity. 14 precautions when seeing these patients in clinic. 3 We anticipate that this would help aid early diagnosis and Although, the exact mechanism of smell and taste mitigate the spread of COVID-19. Therefore, we recommend dysfunction in COVID-19 is not known, there are a few the addition of anosmia, hyposmia, ageusia and dysgeusia to theories out there. Firstly, it could be due to direct damage of the list of criteria for screening of COVID-19. the virus to the olfactory receptors. 41 Secondly, COVID-19 causes to the olfactory epithelium. The The authors earlier set out to assess viral loads and smell and epithelium contains sustentacular cells that structurally taste dysfunction but yielded no results. Although the support sensory neurons, provide nutrients, and maintain the majority used COVID-19 RT-PCR, they were mostly salt and water balance. Therefore during an infection, qualitative and not quantitative. Therefore, there needs to be damage of sustentacular cells could cause a disruption in the studies on quantitative RT-PCR viral loads in association with olfactory epithelium and degradation of sensory neurons anosmia/hyposmia scoring. leading to anosmia. 42 This was supported by a recent study that found that expression of both ACE2 and TMPRSS2 on the In light of suggestions of variation of ACE2, TMPRSS2, protein olfactory epithelium sustencular cells comparable to those S and protein C, further studies should include tissue biopsies observed in lung cells. 42 Thirdly, nasal subepithelial mucosal from the olfactory mucosa to gain a correlation (if any) inflammation could prevent odours from reaching the between the presence of these receptors with olfactory mucosa. 43 hyposmia/anosmia.

Although questionnaires may be quick, a combination of a quantitative test (e.g., University of Pennsylvania Smell LIMITATIONS This review is subject to several limitations. Firstly, most of Identification Test - UPSIT, Sniffin Stick Test) and subjective the studies were retrospective in nature. For the studies that questionnaires with elimination of confounding factors such were prospective, some studies interviewed their patients as chronic rhinosinusitis and nasal polyposis may provide for through the telephone and questionnaires. Therefore, there a more robust study. was a possibility of recall bias regarding the accuracy of symptoms of patients. Lastly, we also recommend the avoidance of nasal endoscopy among patients with smell and taste dysfunction for the first Secondly, there was a lack in formal quantitative two weeks of onset of symptoms in the absence of red flag chemosensory testing and as such may have led to an symptoms, unless necessity dictates so with adequate PPE as overrepresentation of anosmia and ageusia. they could potentially be infected with COVID-19.

Thirdly, smell and taste dysfunction in the COVID-19-positive patients were more likely to respond based on the increased CONCLUSION media reporting of smell loss and/or desire to share their Smell and taste dysfunction has a high incidence in Europe, experience, as compared to the COVID-19-negative patients North America and the Middle East. The incidence is lower in especially when obtaining data from telephone interviews the Asia region. It is a strong risk factor for COVID-19. The and online questionnaires. 15,18 However, the argument symptoms may appear just before, concomitantly, or against the above limitations was that anosmia, hyposmia, immediately after the onset of the general symptoms of ageusia and dysgeusia were not ambiguous symptoms and COVID-19 and usually resolve within one to two weeks. They would likely to be recalled by patients given the lack of these appear predominantly in the younger age group and female symptoms in COVID-19 negative patients. 15 gender. Anosmia and hyposmia usually appear in tandem with ageusia and dysgeusia. These symptoms should be Fourthly, follow-up was reported in only five studies evident added to the list of symptoms when screening for COVID-19. by the median recovery time of smell and taste dysfunction. 11,13-15,25 As such, the incidence of long term smell and taste dysfunction was not known.

Med J Malaysia Vol 75 No 5 September 2020 577 19-A reviewCOVID-final_3-PRIMARY.qxd 9/7/20 8:30 PM Page 578

Review article

9 o

e . 9

t ) r

7 r s

1 5 e a d -

d

r r % m f e f

n g

D e n t 6 o o h o I o r

e o a S P .

t

a t

h

u r o s i o V .

s 5 )

s c N d . i n p f a o i p s

. u a 5

s

a O w i r s t C n o (

z

%

d o i m a a

o h , l C n ) a h l o i i t 9 e 2 e i n

y t y s n t o l s s n a s . n i

s a t r 1 e ( c h n g e c

l i - 1 u u g n m i g t g s o u w c

n S n g

o 9 e e a l i t o

a i D r n u i s ( s i I o o f f r o

c i N g g i

e r i n i e i t t d

d v t r n 9 u V

y m C a a c

e e l

i n d a

t d c r a e 1 t e

e i i e O o d e - o d h d d v n o d t r n n o v m p r v t C n

i f t e c n D n n a s i a

e n m e

p I t

n

s n s a

e

a i a a

t i

e

n

s h s

d

w e

s V t o i 9 m 9 d m i

n e i

d t n n s a e a a s n 1 1 n m s o a r t o i i i h O a r e a - - d e o o t t r o e c o t a i i o t r g C a i

e m l s g r m n p D g D m m

c c n e t

e s I I i i u s p s s o u m n y D e o o g a g h d l

i u m V p p o V o r o o c l p T o

p r r t m s s o r y e t g o r e n n n o i o O e r O a s a r a & o t u u e h s m R A s h p i T m S c C a s b A e i s A p a P r t C l e

r - l d a

a c e e i d c v l x t

a a i i i e f / / r L o n n c m m

d d & 7

1 : n . a d i a 7 n o i : 1 i a . s t m i n 3 s a u a

d r : e o e e a a u / / g n D D n M S M a a n

s s t n ) n o e i ) %

t

d r 8 ) ) l a . l e n % d 0 a 0 d o % % r 6 n 7 7 0 n p ( e e s

> 6 6 e v ( ( ( e m G F o r M F F m o ) c 4

e

5 r 7 )

n : : - / 1 s a

d n n r i n ±

5 n a a a . a d 6 5 e . . a e e e a 4 a e e

g / / 2 6 y 2 s M ( A M n M 5 ( 3 M n t l u s e , r ) e

r r h

e o t ) t s

i f

: f e 1 w e a % m

t b 2 )

3 , o

( . s t : u

g g t g t 2 % c e p n n p n e t 3 7 a i i i i . ( r r r m r s u

b a 7 c u u y c u n / 0 u 6 s S d o n A ( S 1 D D u n

a ) 1 : ) ,

9

a

m : : % ) )

4

i )

) )

% 1 : 0 r r ) ) 6 9

9 1 s . 6 d 3 a 9 e e % )

% % . 7 % i 7

2 ) ) u t t 5 % % e 1 1 e s 8 8 2 t t ( 0 = . = = = ) e . % . 2 i i 2 . v

= ) , u % ) 2 w % . . 5 n n n n g ) 6 5 5

9 ( 2 b b + ( n ( %

e 7 . e

4 5 5

/ / (

,

, , 0 4 . 2 i 2 ( % 1 A % 2 ,

t t & 6 g s 4 ( 4 (

7

( . 1 9 9 l l 9 9

4 3 v ( t

3 (

(

: 5 9 9 : ( l

5

: 1 (

. A 9 1 1 a a 1 1 e R e 7

8 ) 9

a 1

1

- - - - n s s 2 3 4

a u 9 4 i 2

z 3 : - - ) / /

3 7

: i : i O R ( , &

1 1 : s 9 1 D D D t t D s 1 %

a s

: : : :

a D D I I I I

a i ( : S ) ) a a i e I I e e : : i 3 % a u

1 u I

a a a a a i . i s V V V V z

i i i i i e e a a 7 e V 5 V m 3 e R e i i 5 % . % m

u e l

: n m O O O O m s g s : s l s g w w m m m m m 1 O 8 7 0 O s e e p D D ( ( C C C s s C s s s s s s o o u u o o b D 6 0 C C

o g u ( T

T

o o o o e o o e p p l a m p T s s p 1 8 4

e e e e e e f ( e e T n n n n g n n y y y y y g y a & & v v v v v v 8 5 1

& v v n S S + A - A + S A A 2 A A 4 H A + H H S D D + 2 + 5 A H - I +

,

,

, s / d

t d e

. , e

e n r l e e e r n d i e e l i g r &

v v s i e e a z , . e

t i i o a i i r a s s t t r v

t n e e d n t n i t c t c S e C a n e t n n e

w

n n l e e n n . c c D p l e o o e o e t

o

l i i p e p o e i c r o i h h t s s e c y v i e t p r t p l s

o r t p p s o o t d s l a e e g r r m e m h e e e n u o s l l t u t t p u n o t a o r u i o a e e e e n n S S C c Q p M c t i i t r H r q S n

i s a d

p , n S )

n a , ,

y l

e , r r , n

d n t i g u e e a a o i r a n i t g h e h d n u a g m t r a i t j u a e o i e e o h S R C ( N N M K D C W

) )

) 0 . 0 .

l 0 2 l n n n 2 y a 2 0

a

o o o r 0

i i i 0 2 e

t a t 2 t t t

)